3. Anthologies and Anthologists 87 Derives Its Epigrams from the Anthology of Cephalas, Was Compiled During the Reign of Leo VI (886–912)14

Chapter Three ANTHOLOGIES AND ANTHOLOGISTS

Between c. 850 and 950 many Byzantine intellectuals, among them brilliant scholars such as Leo the Philosopher, devoted themselves wholeheartedly to the study of ancient, late antique and contemporary poetry. These hundred years of Byzantine scholarship resulted in the compilation of two major anthologies: the Palatine Anthology (compiled shortly after 944) and the Anthologia Barberina (c. 919). The latter is a collection of Byzantine anacreontics and alphabets, which can be found in Barb. gr. 310 (see below, pp. 123– 128). The former is essentially a copy of an earlier anthology of epigrams put together by Constantine Cephalas at the end of the ninth century. The anthology of Cephalas is not preserved, but we can reconstruct its structure in broad outline with the help of various collections of epigrams that derive from it, either directly or indirectly. Of these collections the Palatine Anthology is by far the most important because it closely resembles the original anthology of Cephalas. The Palatine manuscript1 was written by six different scribes2. These six hands can be divided into two groups: B1, B2 and B3, and J, A1 and A2, respectively. Both groups of hands can be dated approximately to the second quarter of the tenth century: scribes B to c. 920–930, scribes J and A to c. 940– 9503. The oldest part of the manuscript, copied by scribes B1, B2 and B3, comprises the epigrams starting from AP IX, 563 to the end of AP XIV (pp. 453–642),

1 After the Napoleonic wars the Palatine manuscript, with the exception of its last 100- odd pages, was sent back to Heidelberg (Pal. gr. 23); the remainder stayed in Paris (Par. Suppl. gr. 384). For the curious wanderings of the Palatine manuscript, see CAMERON 1993: 178–201. 2 For a thorough description of the manuscript, see J. IRIGOIN, Annuaire de l’ École Pratique des Hautes Études,1975–76. Sect. IV. Sciences Historiques et Philologiques, 281– 295. 3 Thus IRIGOIN (see footnote above), 283–284, and A. DILLER, in: Scripta Turyniana, ed. J. HELLER. Urbana 1974, 520–521. M.L. AGATI, BollClass, t.s., 5 (1984) 42–59, dates both sets of hands a few decades later: scribes B about 940–950 and scribes J and A about 960–970. CAMERON 1993: 99–108 suggests that the two groups of scribes, albeit working in different scriptoria, cooperated in a joint venture under the guidance of the chief editor J; this theory has been refuted by J.-L. VAN DIETEN, BZ 86–87 (1994) 342–362. 84 Part One: Texts and Contexts as well as AP XV, 28–40 (pp. 705–706 and 693–695). Since the outside leaves of the last quaternion, no. 44 (pp. 691–706), were accidentally folded wrong during binding, the original order of the epigrams is as follows: AP XV, 40 and 28–39. The rest of the manuscript (pp. 1–452, 643–692 and 696–704) was written by J, A1 and A2. It contains the first four books of Cephalas’ anthology: AP V, VI, VII and IX, 1–562, plus the introduction to it, AP IV4. It also contains AP VIII (Gregory of Nazianzos’ epitaphs), a book that does not belong to the original Cephalas, but was added to it in the early tenth century5. Before and after the anthology of Cephalas we find various long poems and collections of epigrams. The poems at the beginning of the Palatine manuscript are the following: Nonnos’ Paraphrase of the Gospel according to John (no longer extant due to the loss of seven quires), Paul the Silentiary’s Ekphrasis of the Hagia Sophia and of its Ambo, various dogmatic poems by Gregory of Nazianzos, a collection of Christian epigrams (AP I), Christodoros of Thebes’ Ekphrasis of the Statues in the Zeuxippos (AP II), and a collection of inscriptions found in a temple at Kyzikos (AP III). At the end of the manuscript, after pp. 453–642 written by scribes B, we again find a hotchpotch of various poems: John of Gaza’s Ekphrasis of the World Map in the Winter Baths of Gaza, a collection of epigrams (AP XV, 1–20 and 23), the Hellenistic Technopaegnia (AP XV, 21–22 and 24–27), and the Anacreontea. Then we have the last quaternion (no. 44), the first pages of which were copied by scribe B3; on the remaining pages scribe J copied various poems by Gregory of Nazianzos. There can be little doubt that scribe J is the final redactor of the manuscript. Scribe J supplements lacunas, adds lemmata and ascriptions, and attempts to unite the various parts of the manuscript so that the seams do not show. In his magnificent book on the Greek Anthology, Alan Cameron con- vincingly proved that scribe J is none other than the famous tenth-century poet, Constantine the Rhodian, and demonstrated that the Palatine Anthology was compiled not long after 9446. The so-called Corrector examined the manuscript after it had already been executed, and made a great number of excellent corrections, for which he used an apograph of Cephalas’ anthology made by

4 Incidentally, this also explains the scholion attached to AP IV, 1, stating that the anthology of Cephalas was divided into four categories ™n t/ parönti ptykt5ù: namely, erotic, anathematic, sepulchral and epideictic (=AP V, VI, VII and IXa). By this, scribe J simply means to say that the present volume, copied by himself and scribes A, contains only these four categories. The scholion does not apply to the rest of Cephalas’ anthology, which was copied by scribes B. 5 See CAMERON 1993: 145–146. 6 CAMERON 1993: 108–116 and 300–307. See also P. ORSINI, BollGrott 54 (2000) 425–435, who, for no good reason, questions the validity of Cameron’s arguments. Anthologies and Anthologists 85

Michael Chartophylax7. On various pages of the Palatine manuscript we also detect a number of additional epigrams copied by a twelfth-century scribe, Sp. The structure of the Palatine manuscript is fairly complex. It is reasonable to assume that the manuscript copied by scribes B did not only contain AP IX, 563 – AP XIV and XV, 28–40, but also the preceding books of Cephalas’ anthology. For one reason or another Constantine the Rhodian (scribe J) had obtained only the second part of the B manuscript and, desiring to have the whole Cephalas, ordered scribes A to copy the rest under his guidance. This they did with the utmost diligence. For reasons unknown to us, Constantine the Rhodian separated the last few pages from the rest of the B manuscript by inserting three new quaternions (41–43) containing John of Gaza’s Ekphrasis, the Technopaegnia and the Anacreontea. And since there were still a few pages left blank between the Ekphrasis and the Technopaegnia, he filled these spare pages (pp. 664–668) with various epigrams. Constantine placed the last few pages of the B manuscript at the very end, after quaternions 41–43. These pages originally formed a ternion. Constantine turned it into a quaternion and copied some poems by Gregory of Nazianzos on the pages left blank by scribe B3 and on the pages he had added himself. Although we are greatly indebted to Constantine the Rhodian for his editorial work on the Palatine manuscript, it cannot be denied that Constan- tine was sometimes a somewhat sloppy editor. On the last pages of the manuscript Constantine copied 68 epigrams by Gregory of Nazianzos, apparently unaware of the fact that these same epigrams could be found in AP VIII, a book copied by his fellow scribe A1. Only when the manuscript was already finished and he had begun checking the work of his fellow scribes, did he notice the duplication8. Constantine’s negligence shows most clearly at AP IX, 583– 584, where he failed to notice a major lacuna. If Constantine had checked other manuscripts of Cephalas’ anthology, he could easily have spotted the lacuna, but for one reason or another he did not closely examine the B manuscript in his possession. The exemplar used by scribes B must have missed three or four quaternions between AP IX, 583 and 584 containing some 450 epigrams on works of art. Most of these epigrams can be found in the Planudean Anthology (printed as book XVI, the “Appendix Planudea” (APl 32–387), in modern editions of the Greek Anthology), a few in the so-called syllogae minores, and some others in the Palatine manuscript itself as additions by the twelfth- century scribe Sp (for instance, AP IX, 823–827 and XV, 41–51). The manuscript that scribes B used did not only lack a considerable amount of epigrams, but also a title and a prooemium separating the epideictic epigrams (AP IXa

7 CAMERON 1993: 116–120. 8 CAMERON 1993: 107–108. 86 Part One: Texts and Contexts

= AP IX, 1–583) from the epigrams on works of art (AP IXb = APl 32–387 + some epigrams in the syllogae minores and the additions of Sp + AP IX, 584– 822)9.

** *

Constantine Cephalas

Little is known about Constantine Cephalas. In sources other than the Palatine manuscript he is mentioned only once: as protopapas at the Byzantine court in 91710. The scholia in the Palatine manuscript unfortunately do not supply us with much valuable information about his person or his activities, except for an intriguing note of the Corrector at AP VII, 429: “Cephalas propounded (proeb1leto) this epigram in the school of the New Church in the time of Gregory the Headmaster of blessed memory”. The scholion informs us that Cephalas used to teach at the school of the New Church and that he once lectured on AP VII, 429, a pröblhma that his students had to solve11. In the prooemia attached to AP V–VII, IX–XII and XIV, Cephalas addresses his students directly every time he introduces a new epigrammatic sub-genre: “you should know (…)”, “please notice (…)”, “you may find (…)”. The per- emptory tone and the didactic tenor of these proems leave no doubt that the anthology of Cephalas came into existence in the context of the Byzantine educational system. Cephalas was a junior teacher at the school of the New Church; the headmaster (mag5stzr) was Gregory of Kampsa, whom we know to have compiled a collection of ancient verse inscriptions, which was incorporated in the anthology of Cephalas12. Seeing that the New Church was inaugurated in 88013, the anthology of Cephalas was published at the earliest in the 880s, if not later. But apparently not much later, for the Sylloge Euphemiana, which

9 See LAUXTERMANN 1998c: 526–529. 10 See Theoph. Cont., 388–389 and Georg. Cont., 881. Keóal@ß is a nickname and means “Bighead”, see Georg. Cont., 820. 11 See CAMERON 1993: 109–110 and 137. For riddles as part of the Byzantine school curriculum, see N.G. WILSON, Scholars of Byzantium. London 1983, 23. 12 For Gregory of Kampsa and his collection of verse inscriptions, see pp. 72–74. For information on Byzantine schools and teachers, see LEMERLE 1971: 242–266 and SPECK 1974a: 29–73 (for Cephalas, see esp. p. 61, n. 28). 13 For the New Church, see P. MAGDALINO, JÖB 37 (1987) 51–64. The school of the New Church seems to have existed only for a short while, seeing that the letters of the Anonymous Professor, dating from 920–940, inform us that the clergy of the New Church sent their protégés to his school, see LEMERLE 1971: 206, n. 3. Anthologies and Anthologists 87 derives its epigrams from the anthology of Cephalas, was compiled during the reign of Leo VI (886–912)14. Moreover, the collection of epigrams at the end of the B manuscript (AP XV, 28–40) provides an important chronological clue that has gone unnoticed. The original lemma attached to AP XV, 32 reads: “by Arethas the Deacon”, to which scribe J added in the late 940s: “who also became archbishop of Caesarea in Cappadocia”15. This clearly indicates that the original lemma was written when Arethas had not yet become archbishop: that is, before 902. Taken in conjunction, the above data suggest that the anthology of Cephalas dates from the last decade of the ninth century. The anthology of Cephalas consisted of the following nine sections: (1) erotic (AP V), (2) anathematic (AP VI), (3) sepulchral (AP VII), (4) epideictic (AP IXa), (5) on works of art (AP IXb), (6) protreptic (AP X), (7) bacchic (AP XIa), (8) scoptic (AP XIb) and (9) pederastic (AP XII). It was followed by a collection of epigrams in unusual metres (AP XIII) and by a collection of riddles, mathematical problems and oracles (AP XIV). At the beginning of his anthology Cephalas placed the ancient prefaces in verse attached to the Gar- land of Meleager, the Garland of Philip and the Cycle of Agathias (AP IV)16. The contents of the original Cephalan compilation do not fully correspond with the modern concept of an “anthology”, a collection of poems put together with the objective to bring like to like. It is worth noticing that Cephalas did not restrict his collection merely to epigrams, but also included two long poems that are certainly not epigrammatic, Nonnos’ Paraphrase and Christodoros of Thebes’ Ekphrasis (AP II). Likewise, Constantine the Rhodian added non- epigrammatic material at the end of the Palatine manuscript: John of Gaza’s Ekphrasis, the Technopaegnia and the Anacreontea. It is not known whether it was Cephalas or Constantine the Rhodian to whom we owe Paul the Silen- tiary’s Ekphrasis and Gregory of Nazianzos’ theological poems (found at the beginning of the Palatine manuscript), but it does not really matter. As I pointed out in the second chapter (pp. 68–69), Byzantine manuscripts may contain a hotchpotch of various kinds of poetry, varying from short epigrams to long poems. The medieval approach to poetry is not as rigid and priggish as that of the moderns, and it is certainly not based on any considerations of genre; anything of interest may be copied and, judging by the contents of Byzantine manuscripts, actually was copied. It is therefore hardly surprising that we find non-epigrammatic texts before and after the actual anthology. Cephalas and Constantine the Rhodian simply followed the editorial practice of their time.

14 See CAMERON 1993: 254–256. 15 See CAMERON 1993: 313. 16 See CAMERON 1993: 121–159. 88 Part One: Texts and Contexts

As for the collections of epigrams found before and after the anthology of Cephalas, it is not always clear who put them there: Cephalas himself, Constan- tine the Rhodian or someone else. The collection of Christian epigrams in AP I was certainly to be found in Cephalas, as will be shown in the next section. The short sylloge at the end of the B manuscript, AP XV, 28–40, may perhaps have been part of the original Cephalas, but I am inclined to think that it is a later addition to the anthology of Cephalas (see pp. 107–108). Constantine the Rhodian’s own contribution to the Greek Anthology is the small sylloge of epigrams copied between John of Gaza’s Ekphrasis and the Technopaegnia (see pp. 116–118). For his anthology of epigrams (AP IV–VII and IX–XIV) Cephalas made use of several sources, of which the five most important are: the Garland of Meleager (1st cent. BC), the Garland of Philip (1st cent. AD), the Anthologion of Diogenian (2nd cent.), the Palladas Sylloge (6th cent.) and the Cycle of Agathias (c. 567)17. Cephalas’ anthology did not contain contemporary epigrams. The only exceptions are Cephalas’ own preface to the book of erotic epigrams (AP V, 1), and some epigrams by Leo the Philosopher and Theophanes the Grammarian (see pp. 100–101 and 104–105). There can be no doubt that Cephalas’ main objective in compiling his anthology was to rescue from oblivion the epigrammatic legacy of the ancients. Cephalas’ scholarly pursuits are not “antiquarian” or “encyclopedic”, as some maintain18, but bear proof of the revived interest in classical literature in the ninth and tenth centuries. This cultural revival manifests itself in the many manuscripts copied in this period as well as in the direct quotations or indirect literary allusions with which contemporary writings are replete. Since he was an intelligent, though sometimes absent-minded editor, Cephalas understood that his task went beyond the limits of mere copying, but involved above all a scholarly approach in sorting out the material at his disposal. That is why he did not copy the epigrams in exactly the same order as he found them in his manifold sources, but attempted to rearrange them (not always successfully) according to genre. His system of classification is essentially the same as that of Agathias, with the addition of two new categories: protreptic and pederastic19. Cephalas’ working

17 For the Garlands of Meleager and Philip, see CAMERON 1993: 49–65. For the Anthologion of Diogenian, see P. SAKOLOWSKI, De Anthologia Palatina quaestiones. Leipzig 1893 and CAMERON 1993: 86–90. For the Palladas Sylloge, see A. FRANKE, De Pallada epigramma- tographo. Leipzig 1899 and LAUXTERMANN 1997. For the Cycle of Agathias, see MATTSON 1942 and A. & AV. CAMERON, JHSt 86 (1966) 6–25. 18 LEMERLE 1971: 268 calls the anthology of Cephalas “une encyclopédie épigrammatique”. Lemerle’s “curiosa affermazione” was refuted by P. ODORICO, BZ 83 (1990) 5–6 and CAMERON 1993: 334–335. 19 Cephalas divided Agathias’ fourth category, “on the devious paths of life, etc.”, into two “books”: epideictic (AP IXa) and protreptic (AP X), probably because one of the Anthologies and Anthologists 89 methods are not entirely clear to us; he may have used file cards in order to avoid duplications and he may have had some assistance from fellow scholars, such as Gregory of Kampsa and the anonymous ™klex1menoß whom the Correc- tor criticizes at AP IX, 16 for his stupidity. Cephalas has not been spared the scorn of modern schoolmasters, who crudely accuse him of aggravating negligence, ignorance and sloppiness. But taking into account the size of the material he was working with and the number of mistakes he could have made, but did not make, these criticisms hardly seem justified. In fact, the fortunes or mishaps of Cephalas’ scholarly work should be judged, if at all, against the background of other ninth- and tenth-century compilations, such as the corpus of short poems attributed to Theognis or the various gnomologies compiled in this period20. Short texts need to be rearranged in such a manner that an anthology or gnomology appears to assume a logical, almost natural coherence; but this seemingly coherent system of classification is, of course, the work of an individual anthologist, who superimposes his own interpretation of, and adds signification to, the texts he is rearranging. In the following, I shall try to characterize the various anthologists who contributed to the Greek Anthology.

** *

A Collection of Christian Epigrams: AP I

The Christian epigrams in AP I21 were copied by scribes J and A1, who apparently cooperated and wrote the text in shifts. Taking into account the scribal error at AP I, 116, it is beyond doubt that the collection of Christian epigrams was not compiled by scribe J himself, but already existed in manuscript form. On pp. 61–62 we find the following epigrams: AP I, 115; 116. 1–2 (with an asterisk indicating that it should be deleted); 116. 3–4; and 30 (dupli- cated here). The text of AP I, 116. 1–2 should indeed have been deleted in modern editions. It begins with the first words of I, 30 and ends with the last words of I, 116. 3–4. Here we have a classic example of haplography, caused by

sources he used, the Palladas Sylloge, contained a great number of protreptic epigrams. Cephalas added the category of paederastica (for obvious reasons absent from the Cycle of Agathias) because of the many epigrams of this kind found in one of his sources, the Boyish Muse of Strato of Sardis. See LAUXTERMANN 1998c: 527–528 and 535–536. 20 For the late ninth-century edition of Theognis, see M.L. WEST, Studies in Greek Elegy and Iambus. Berlin–New York 1974, 44–45. For ninth- and tenth-century gnomologies, see ODORICO 1986: 3–28. 21 For studies on AP I, see especially WALTZ 1925, BAUER 1960–1961 and BALDWIN 1996. 90 Part One: Texts and Contexts the presence of the word 4óqiton both in I, 30 and in I, 116. 3–4. Scribe A made up for his mistake by rewriting AP I, 116. 3–4 in its original form, while scribe J, taking over on the next page, wrote down the text of AP I, 30. The original sequence of epigrams in the exemplar they were copying must have been as follows: AP I, 115; I, 30; and I, 116. 3–4. This also explains the heading attached to AP I, 116: “on the same”, i.e., “on Christ” – the subject matter, not of AP I, 115, but of AP I, 30. The collection of Christian epigrams is not a later addition to the anthology of Cephalas, as most scholars seem to believe, but forms part of the original Cephalas. First of all, as Alan Cameron observed, at least four epigrams in AP I (nos. 33–36) derive from the Cycle of Agathias22. It seems very unlikely that Cephalas, while thumbing through his exemplar of the Cycle, would have skipped these beautiful epigrams only because they deal with archangels instead of pagan deities. In fact, the mere suggestion would question the ethics of the very person who was to become protopapas at the Byzantine court. Secondly, the collection of Christian epigrams was also to be found in two independent copies of Cephalas’ anthology: the Cephalan source used by the Souda for the numerous epigrams it quotes, and the apograph made by Michael Chartophylax and checked by the Corrector. The Souda quotes a few verses from epigrams in AP I23, and the Corrector makes no less than fifteen corrections in the text of the Palatine manuscript. Most of these corrections are insignificant and may have been the Corrector’s own conjectures, but the excellent emendations: l7ssan instead of l8qhn (AP I, 10. 72) and än5acon instead of än5scon (AP I, 92. 3), indicate that the Corrector had a better text in front of him24. Thus there were at least three tenth-century manuscripts com- bining the collection of Christian epigrams with the anthology of Cephalas: the Palatine manuscript itself, Michael Chartophylax’ apograph and the manuscript used by the redactors of the Souda. Thirdly, AP I contains a great number of verse inscriptions. As one would expect, most of these verse inscriptions were copied in Constantinople: AP I, 1–18, 96–98, 104, 106–107, 109–114 and 120–121; but the epigraphical survey also included other Byzantine cities, such as Ephesus, Caesarea and Cyzicus: AP I, 50, 91, 92–93, 95 and 103. As Gregory of Kampsa is known to have visited these cities for his collection of verse inscriptions, it is very likely that he is the epigrapher who contributed to what was to become AP I.

22 CAMERON 1993: 152–158. See also the interesting study by P. SPECK in: Varia II (Poik5la Byfantin1 6). Bonn 1987, 357–362. BALDWIN 1996: 101–102 is not entirely convinced by Cameron’s arguments. 23 See CAMERON 1993: 151. 24 For the 15 corrections, see STADTMÜLLER 1894–1906: ad locum, AP I, 10. 51; 10. 72; 19. 3; 63. 2; 65. 1; 66. 1; 66. 2; 67. 1; 85. 1; 86. 2; 90. 1; 92. 3; 94. 6; 98. 4; and 116. 1. Anthologies and Anthologists 91

Furthermore, there are also some interesting lemmata and scholia in AP I that indirectly indicate that the collection of Christian epigrams must have been compiled by Cephalas himself. AP I, 106–107, are two verse inscriptions celebrating the decoration of the Chrysotriklinos commissioned by Michael III; they date from 856–866. AP I, 109–114, too, are verse inscriptions; they were found in the church of the Virgin of the Source, which was decorated by Basil I and his sons Constantine and Leo in the years 870–879. Although verse inscriptions are destined by their very nature to remain anonymous, the anthologist of AP I duly records the names of the poets who wrote the above epigrams: a certain Mazarenos (AP I, 106–107)25 and an equally obscure school- master, Ignatios the Headmaster (AP I, 109–114)26. From this we may infer that the anthologist had firsthand information on the two poets and their literary achievements in the 860s and 870s. Otherwise, how could he have known which poets out of many possible candidates had been commissioned to compose the anonymous verses he found inscribed in the Chrysotriklinos and the church of the Pege? It is reasonable to assume that the well-informed source used by Cephalas was none other than the collection of verse inscriptions compiled by Gregory of Kampsa. Gregory lived in exactly the same period as the two poets and there can be little doubt that he must have personally known at least Ignatios the Headmaster, a colleague of his. At AP I, 122 we find another name of a member of the circle of Cephalas: Michael Chartophylax, whose personal copy of Cephalas’ anthology was used by the Corrector. At AP I, 10, a long verse inscription found in the church of St. Polyeuktos, we find the following curious scholion: m6noysin, 4riste, p1nta m6cri t‰ß s8meron Çtesi pentakos5oiß. Since the church of St. Polyeuktos was built by Anicia Juliana between 524 and 52727, the scholion appears to err in its arithmetic. However, if one follows the inaccurate dating provided by the Patria, according to which Anicia was the daughter of Valentinianus and the sister-in-law of Theodosius the Great28, we arrive at a date in the late ninth century29. The lemma attached

25 For this name, see WALTZ 1925: 321–322, who suggests that the poet, or his family, came from a place called “Mazara”. 26 Ignatios the Headmaster should not be confused with Ignatios the Deacon (born c. 780, died c. 850): pace MAKRIS 1997: 10 and 12; see WOLSKA-CONUS 1970: 357–359 and MANGO 1997: 13. As for Ignatios’ title: mag5stzr t0n grammatik0n, see the name of the school located in the Orphanage of St. Paul (s. XI–XII): scolë t0n grammatik0n, and the title of one of its principals (Basil Pediadites): maÀstzr t‰ß scol‰ß t0n grammatik0n, see P. LEMERLE, Cinq études sur le XIe siècle. Paris 1977, 233–234. 27 C. MANGO & I. ŠEVCENKO, DOP 15 (1961) 243–247. 28 Ed. PREGER 1901–07: 57. See P. MAAS, Hermes 48 (1913) 296, n. 2 and CAMERON 1993: 114. BALDWIN 1996: 98 finds it hard to believe that “the scribe (would have been) this obtuse”. 29 Some fifty years later, scribe J tried to bring the scholion up to date by adding kaò …konta, but afterwards erased his own addition. 92 Part One: Texts and Contexts to AP I, 7 states that an amount of money was found hidden in the church of St. Theodore. The same story is told in more detail by the Patria, from which we learn that the miraculous discovery of the treasure took place during the reign of Leo VI30. Taken in conjunction, the above data can lead to one conclusion only: the collection of Christian epigrams was compiled at the end of the ninth century in the scholarly ambience of Cephalas. The collection of Christian epigrams is of great interest to art historians, since it provides abundant information on Byzantine monuments that either no longer exist or remain only as sad ruins of glory and magnificence lost for ever. Two of the many verse inscriptions in AP I are still partially extant. Some traces of AP I, 1 can still be seen in situ: on the bema arch of the Hagia Sophia, above the famous mosaic depicting the Holy Virgin with Child31. Recent excavations at Saraçhane have brought to light a few fragments of AP I, 10, an encomiastic ekphrasis of no less than 76 verses which, despite its non-epigram- matical length, was actually inscribed on the walls of the church of St. Poly- euktos32. It is not always clear where Cephalas found the epigraphic material he used in his anthology. Did he read the Polyeuktos ekphrasis in a literary source or did Gregory of Kampsa provide him with a copy of the verse inscription? Neither of these two possibilities can be ruled out in view of AP I, 99 and AP I, 120–121. AP I, 99 is a genuine verse inscription, but Cephalas derived it from a literary source, the Life of Daniel the Stylite33. AP I, 120 and 121 are two epigrams on the Blachernai church, which we know to have been written by George of Pisidia. Although one would expect that Cephalas culled these epigrams from the collection of Pisides’ poems, the fact that the lemma attached to AP I, 120–121 notes their provenance, but not their author, strongly suggests that the two epigrams were copied in situ. The fate of AP I, 92 at the hands of modern editors is somewhat bizarre. This epigram can be found in standard editions of Gregory of Nazianzos (I, 1, 28), even though it is a dubious attribution resting on the slender evidence of two manuscripts, Par. gr. 1220 and Monac. gr. 416, where the epigram is written at the end of various Grego- riana. In the former manuscript the epigram is followed by Ignatios the

30 Ed. PREGER 1901–07: 30. See G. DAGRON, Constantinople imaginaire. Études sur le recueil des Patria. Paris 1984, 155–156 and n. 116–117, MANGO 1986: 25–28, and BALD- WIN 1996: 97. 31 See E.M. ANTONIADIS, èEkórasiß t‰ß ^Ag5aß Soó5aß. Leipzig–Athens 1907–1909, III, 29– 31, and MERCATI 1922a: 280–282. 32 See C. MANGO & I. ŠEVCENKO, DOP 15 (1961) 243–247 and R. HARRISON, Excavations at Saraçhane in Istanbul. Princeton 1986, I, 3–10 and 405–420. See also P. SPECK, in: Varia III (Poik5la Byfantin1 11). Bonn 1991, 133–147, and C.L. CONNOR, Byz 79 (1999) 479– 527. 33 See CAMERON 1982: 247–252. Anthologies and Anthologists 93

Deacon’s anacreontic, in the latter by Ignatios’ anacreontic and Pisides’ De Vanitate Vitae, vv. 41–56: these two poems, too, have been included in modern editions of Gregory of Nazianzos as if they were his (Epit. 129 and I, 2, 18)34. In Monac. gr. 416 AP I, 92 is not attributed to Gregory of Nazianzos, but to “Basil the Great”35. This ascription is also incorrect. But it implicitly indicates from which source the two Gregorian manuscripts ultimately derive the epigram: the Greek Anthology, where it bears the following title: “in Caesarea in the church of St. Basil”. AP I, 92 is in fact a verse inscription. The verse inscription still exists (unfortunately, in a rather garbled version), not in Caesarea itself, but in the nearby village of Sinassos, at the entrance of the church of the Holy Apostles, where it accompanies a tenth-century fresco depicting Pentecost36. The epigram describes the miraculous intervention of Jesus Christ on the lake of Galilee. Its didactic purpose is to show the two natures of Christ. While the waters rage He sleeps like any other human being, but when He awakes He shows His divine nature by immediately calming the storm. The epigram would certainly have appealed to the pious monks of Cappadocia because of its iconophile emphasis on the two natures of Christ, but it is not entirely clear why they had it inscribed below a picture of Pentecost. To return to our subject, however, it is reasonable to assume that the Cappa- docian monks copied the epigram in Caesarea, where it was inscribed in the church of St. Basil. AP I, 92 is a genuine verse inscription, which ended up in Par. gr. 1220 and Monac. gr. 416 via the Greek Anthology. And thus an anonymous verse inscription became a literary epigram supposedly written by Basil the Great or, if we are to believe modern scholars, Gregory of Nazianzos. Verse inscriptions can be given approximate dates if they mention emperors or other prominent individuals, but metre and language are equally instru- mental in assessing the probable date of a poem. Take for instance AP I, 105, “on Eudokia, Wife of the Emperor Theodosius”, an epigram on a fresco or mosaic that depicted Eudokia venerating the Holy Sepulchre. Fifth-century, one would say a priori. But the metre, regular Byzantine dodecasyllables, obviously militates against such a dating. The verses cannot have been written before c. 600, and may even have been written much later, say in the ninth century. Do poem and picture perhaps form an indirect homage to the Empress Theodora, who showed her piety by restoring the cult of icons and

34 See H.M. WERHAHN, in: Studia Patristica VII. Berlin 1966, 340–342. See also MERCATI 1908: 3–6, GONNELLI 1991: 120–121, and LAUXTERMANN 2003b. 35 As was duly noted by H.M. WERHAHN, in: Bibliotheca docet. Festschrift C. Wehmer. Amsterdam 1963, 342–344, who nonetheless avers that “aus inneren Gründen (…) an die Verfasserschaft tatsächlich zu denken ist”. 36 See H. GRÉGOIRE, Revue de l’ instruction publique en Belgique 52 (1909) 164–166. 94 Part One: Texts and Contexts could thus be presented as a spiritual pilgrim? The collection of distichs at AP I, 37–89 comprises an epigram cycle dating from c. 600: nos. 37–49 and 52– 77, to which Cephalas added various late antique and Byzantine epigrams (nos. 50–51 and 78–89)37. Other epigrams in AP I cannot be dated, such as nos. 104 and 108: probably early Byzantine, but possibly written after 600. Generally, a certain chronological order may be detected in the arrangement of the epigrams. Book AP I has a tripartite structure: 1–36, 37–89 and 90–123, designed to create a mirror effect whereby beginning and end appear to correspond, with the collection of distichs at AP I, 37–89 in the middle. The first and the last parts contain a mixture of verse inscriptions and literary epigrams, but whereas the first 36 epigrams date from late antiquity (with the noteworthy exception of AP I, 1), most of the epigrams at the end of AP I were written after c. 600. Since the spheres of the sacred and the profane intermingle in Byzantium and since God is never far away from the everyday experience of the Byzan- tines, the notion of a “Christian” epigram is in itself utterly unchristian, for it presupposes that there may exist another conceptual world lying beyond the horizons of Christendom. It is for this reason that Byzantine authors hardly ever specify that their literary works should be viewed as the products of a typically Christian ideology. Seen from the perspective of ninth-century Byzantium, the question whether a contemporary epigram is “Christian” or not is totally irrelevant. Of course, there had once been a world that had not known the blessings of Christianity, but was infested with uncanny supersti- tions, pagan cults and lascivious fantasies. That was the world of the Hellenes, about whom the Byzantines learnt at school. Although classical schooling was valued highly in ninth-century Byzantium, if only because it secured social prestige by distinguishing the man of letters from his less educated peers, there was still a psychological barrier to be crossed: a mental watershed between Byzantium and Hellenism, between “us” and “them”. Only in opposition to what is viewed as alien, not “ours”, does the definition of a Christian epigram assume relevance, but since no Byzantine scholar before Cephalas seems to have given much thought to the problem, he had some difficulties in demarcat- ing and outlining the domain of what constitutes a proper Christian epigram. Most of the epigrams in AP I deal with churches, religious images and arte- facts; the remaining are personal prayers, dogmatic poems and book epigrams on Christian literary works. Though there can be little doubt that these epigrams are rightly labelled “Christian”, Cephalas was not as consistent as one perhaps would have liked, for in AP IXb, the section dealing with works of art, we find a number of epigrams that are clearly Christian and should therefore

37 See Appendix X, pp. 357–361. Anthologies and Anthologists 95 have been put in AP I: AP IX, 615, 787, 806–807 and 817–819. But this type of misclassification is actually very common in the anthology of Cephalas; in fact, it is typical of Cephalas to forget or to neglect his original design. Howev- er, it is rather surprising that Cephalas excluded epitaphs from his collection of Christian epigrams. Whereas AP I does not contain any epitaphs, we find in the section of sepulchral epigrams no less than seven epitaphs that are undoubtedly Christian: AP VII, 667, 679–680, 689 and VIII, 138. Take for instance VII, 689: “Here Apellianus, most excellent of men, left his body, depos- iting his soul in the hands of Christ”39. Perhaps Cephalas considered a poem like this inappropriate for his collection of Christian epigrams because it honours a specific individual at a certain point in space and time, and thus forms a memorial of little significance compared to God’s everlasting omnipresence. But there are scores of dedicatory epigrams in AP I that, seen in the light of eternity, are as much a product of their time as the Christian epitaphs. So, why did Cephalas not include epitaphs in his collection of Christian epigrams? There is no answer to this question, but it clearly indicates that an epigram with a Christian subject is not necessarily a Christian epigram, at least not according to Cephalas. Although AP I properly speaking does not belong to Cephalas’ anthology of Hellenistic, early Roman and late antique epigrams (AP IV–VII and IX– XIV), it directly owes its existence to it. Cephalas decided to compile the collection of Christian epigrams as a defensive measure to clear himself before- hand of any suspicions of “paganism” that might be aroused by the “pagan” contents of his anthology. Part of this strategy was to begin the collection with an iconophile statement of faith: the famous verse inscription on the bema arch of the Hagia Sophia, above the splendid apse mosaic depicting the Holy Virgin with Child (AP I, 1)40. The date for the apse mosaic and consequently its verse inscription is 867, the year in which Patriarch Photios delivered a magnificent, but rather abstruse homily on the mosaic and its pictorial meaning41. We may

38 AP VIII, 1 belongs to AP VII, not to the collection of epitaphs by Gregory of Nazianzos in AP VIII. We owe this misclassification to the editio princeps of the Palatine Anthology. 39 The translation is that of PATON 1918 (as are all the translations from the Greek Anthology in the following). 40 The lemma attached to AP I, 1 states that the epigram was inscribed eœß tñ kibo¯rin. BALDWIN 1996: 97 assumes that the word kibo7rion refers to the “cupola”: so do I, but it must be said that the word normally indicates the “baldachin”. P. SPECK, in: Varia II (Poik5la Byfantin1 6). Bonn 1987, 285–312, suggests that the epigram was originally to be found on the baldachin (built shortly after 843) and that it was afterwards re-used for the apse decoration of 867. 41 See C. MANGO, The Homilies of Photios Patriarch of Constantinople. Cambridge, Massa- chusetts 1958, 282–286, and C. MANGO & E.J.W. HAWKINS, DOP 19 (1965) 113–151. 96 Part One: Texts and Contexts not know the name of the poet who wrote AP I, 1, but it is reasonable to assume that the verse inscription bears out the ideas of the person who commissioned it, the patriarch himself. The text of the epigram reads as follows: “The images that the heretics took down from here, our pious sovereigns replaced”. As the original sixth-century decoration of the Hagia Sophia did not include any figural representations, we must conclude that Photios either lied on purpose or did not care much about historical truth. Whether Photios rewrote history and distorted the facts intentionally or not, the message of the verse inscription and the mosaic itself is very clear: iconophily is back in town. That is, with a considerable delay of some 24 years, for the cult of the icons had already been restored in 843. The verse inscription emphasizes the orthodoxy of the reigning emperors by cleverly postponing the word p1lin, so that it indicates not only that the sovereigns replaced the holy images, but also that these emperors were pious again, in contrast to the hideous iconoclasts who had ruled before them. Whereas all other ninth-century epigrams can be found in the last part of the collection (AP I, 90–123), Cephalas placed the Hagia Sophia verse inscription right at the beginning. By putting it there, he obviously intended to make clear from the start that his personal religious views were above suspicion. Cephalas must have felt compelled to declare publicly his “orthodoxy” out of fear that people might think that he sympathized with the unorthodox contents of his anthology. To compile an anthology of ancient epigrams was in itself not objectionable, but it had to be done cautiously so as not to arouse suspicions. In Byzantium the classical heritage is usually approached from the narrow angle of utilitarianism: that is to say, the study of ancient literature is a laudable pursuit only if it serves the aim of acquiring stylistic skills necessary for the composition of Byzantine literary works. It is not so much the content as the varnish of things old that the Byzantines were supposed to value when they read Homer, Euripides or Plato. But since form and content are interre- lated, to involve oneself with the ancients could be quite hazardous. And indeed, some Byzantine intellectuals, such as Leo the Philosopher and Leo Choirosphaktes, were accused of indulging in the ambiguous beauty of classical literature with far too much zeal. Since the ancient gods were dead and no one believed in them any more, there was no real danger there; but what was particularly offensive to the Byzantines, were sexually explicit texts. This explains the cautious tone of Cephalas in the prefaces to the erotic and the pederastic epigrams. The paederastica in AP XII are introduced as follows: “What kind of man should I be (…) if I were to conceal the Boyish Muse of Strato of Sardis, which he used to recite to those about him in sport, taking personal delight in the diction of the epigrams, not in their meaning. Apply yourself then to what follows, for ‘in dances’, as the tragic poet says, ‘a chaste woman will not be corrupted’.” If we are to believe Cephalas, Strato of Sardis Anthologies and Anthologists 97 was not genuinely interested in boys, but wrote his epigrams only to show off his literary talents “in sport” Implicitly, we are told not to pay attention to what is said, but rather to how it is said. The preface to AP V tells us how we are to interpret the erotic epigrams: “Warming the hearts of youth with learned fervour, I will make Love the beginning of my discourse, for it is Eros who lights the torch for youth”(AP V, 1). Here the conceptualized figure of Eros is not unlike the winged creature of Plato guiding the intellectual soul into the spheres of pure contemplation. Cephalas’ students are admonished not to think of physical love, but to abstract themselves from profane thoughts by way of an intellectual process, “learned fervour”. Needless to say, this is pure hypocrisy. The problem for Cephalas was how to sell his product. Of course, he could have skipped the “pornographic”epigrams, as did Planudes, but his aim was to give a representative sample of the ancient epigrammatic art, including the erotica and the paederastica. Although he was well aware of the effect erotic epigrams might have on the reader, he attempted to present ancient eroticism as a quite innocent pastime. The erotic epigrams were to be read merely as exercises in the art of literary discourse, as magnificent words without sub- stance. Still, Cephalas had good reason to doubt that his idea of a textual labyrinth of words referring to other words, and not to some obscene reality, would be embraced without protest by all the readers of his anthology. Know- ing that he easily could be misunderstood despite the priggish prefaces to the two books of erotic epigrams, he felt obliged to pay lip service to orthodox fundamentalists by adding a collection of Christian epigrams. This is also illustrated by Cephalas’ preface to the collection of Christian epigrams: t2 t0n Cristian0n protet1cqz eJseb‰ te kaò qe¦a ™pigr1mmata kÌn oW æEllhneß äpar6skzntai, “Let the pious and godly epigrams of the Christians take precedence, even if the Hellenes are displeased”. The verb protet1cqz is deliberately ambiguous in this context. It indicates not only that the collection of Christian epigrams is placed before the epigrams of the Hellenes (AP IV–VII and IX–XIV), but also that it takes the place of honour. The epigrams of the Christians deservedly rank first because they are Christian – which is a circular argument, of course, but one indicative of the dire straits Cephalas found himself in. He risked being stigmatized as a Hellene himself for publishing an anthology of pagan epigrams. Cephalas obviously felt the need to deny overtly any inclination towards “Hellenism”. The introduction to AP I and the Hagia Sophia epigram with which AP I begins, bear out the same unequivocal message: “I, Cephalas, have nothing to do with the Hellenes, I am really not one of them”. In ninth-century Byzantium all sorts of people were branded æEllhn: iconoclasts, intellectuals, political opponents, and so forth42. There is no need

42 See I. ROCHOW, in: Paganism in the Later Roman Empire and in Byzantium, ed. M. SALAMON. Krakow 1991, 133–156. 98 Part One: Texts and Contexts to take these charges of paganism seriously. But to be victimized in such a manner was most certainly a quite serious matter for those who were being accused of supporting pagan ideas. Cephalas’ petty fears are therefore quite understandable. In fact, seeing what had happened to one of the anthologists of the previous generation, Cephalas had good reason to be afraid.

** *

Leo the Philosopher, Constantine the Sicilian & Theophanes the Grammarian

The Apology of Constantine the Sicilian43 provides an interesting parallel to the defiant words of Cephalas’ prooemium: “This is the worthy plea for a worthy cause, which I, the patricide of an impious teacher, piously put on record, even if the Hellenes may fret with anger and rage in words along with the Telchines”44. The Apology is a very curious text. In it, Constantine the Sicilian tries to defend himself against accusations of having shown a complete lack of piety towards his recently deceased teacher, Leo the Philosopher, when he publicly denounced him as a pagan. By good fortune we also possess the text of the very poem that Constantine’s contemporaries found so repulsive: the Psogos45. It is indeed a sort of spiritual patricide. Constantine heaps a load of bizarre allegations upon his former teacher. Leo did not believe in the triune Godhead of the Christians, but worshipped the ancient gods: lecherous Zeus married to Hera but always fooling around with his paramours, and all those other ridiculous divinities of whom Homer sings the praises. Now that Leo is dead and buried, Constantine wishes him a pleasant stay in hell where he may be punished together with those cursed Hellenes whom he so much admired: Socrates, Plato, Aristotle, Chrysippus, Epicurus, Proclus, Euclid, Ptolemy, Homer, Hesiod and Aratus. Constantine regrets dearly that he discovered the true nature of Leo’s teachings only when it was already too late; but now that he has seen the light, he cannot but tell the world what his former master was really like. That is why he repeats his allegations in the Apology, adding some new damning evidence and declaring his adamant faith in Christianity with the fervour of a newly converted. Reading the two poems, the Psogos and the

43 The lemma attached to the poem should be emendated into: äpolog5a [Kznstant5noy kat2] L6ontoß Uilosöóoy, kaq\ Ùn Cristñn mên s6bei, t2 ^Ell8nzn dê óayl5fei, as MERCATI 1923–25: 235, n. 1, demonstrated. For the identification of the author, see LAUXTERMANN 1999a: 164–166. 44 Ed. SPADARO 1971: 201, vv. 31–35. 45 SPADARO 1971: 198–199. Anthologies and Anthologists 99

Apology, we may understand what Lemerle meant when he wrote: “Nous ne serions pas trop surpris que l’ auteur de ces deux pièces eût l’ esprit un peu dérangé”46. However, although one might question Constantine’s ethics, his splendid style and fine rhetoric clearly show that Leo the Philosopher’s lessons in the art of literary discourse had not been wasted on him. In fact, despite Constantine’s sincere regrets, his literary works undoubtedly bear the marks of his apprenticeship with Leo the Philosopher and the classicistic movement, of which Leo had been the leading figure until the moment of his death (shortly after 869)47. Leo the Philosopher’s unreserved devotion to the ancients and their legacy deeply influenced the generation that came of age in the years 840– 870 and studied at his school at the Magnaura48. Constantine the Sicilian was one of them. He himself had once rallied to Leo’s ideal of an enlightened hellenism. This also explains the bitter tone of the Psogos and the Apology, for Constantine attacked what had once been dear to him and, in the process, had to deny his former self. Leo the Philosopher and his students were interested in just about anything, ranging from the liberal arts to philosophy, mathematics, astronomy and natural sciences. One aspect of their various scholarly pursuits appears to be entirely unknown: namely, collecting and anthologizing ancient epigrams. None of these anthologies, except for the Parisian Collection of Paederastica, has been preserved; but if one studies the text history of the Greek Anthology attentively, there is ample evidence to prove that Cephalas followed in the footsteps of an earlier generation of scholars, whose work he incorporated in his own anthology. The final editor of the Palatine Anthology, Constantine the Rhodian, was apparently aware of Cephalas’ debt to these scholars, for at the end of his manuscript, where we find a small sylloge by his hand (see below, pp. 116–117), he indirectly paid homage to their scholarly work. There we find four poems by four ninth-century scholars: Michael Chartophylax (the scholar whose personal apograph of Cephalas’ anthology was used by the Corrector) and three members of the circle of Leo the Philosopher. In AP XV, 12 Leo the Philosopher, nicknamed Ö æEllhn, expresses his belief as a true Epicurean that

46 LEMERLE 1971: 175. 47 On Leo the Philosopher, see the brilliant essay by LEMERLE 1971: 148–176. See also N.G. WILSON, Scholars of Byzantium. London 1983, 79–84, ALPERS 1988: 353–359, V. KATSA- ROS, in: Science in Western and Eastern Civilization in Carolingian Times, ed. P.L. BUTZER & D. LOHRMANN. Basel 1993, 383–398, CH. ANGELIDI, in: EJvyc5a. Mélanges offerts à H. Ahrweiler. Paris 1998, 1–17, and J. HERRIN, Dialogos 6 (1999) 27–31. 48 Theoph. Cont. 185 and 192. See LEMERLE 1971: 158–160. The Magnaura school opened its gates shortly after 843: see SPECK 1974: 4–7. Whether it already existed during the reign of Theophilos in a different form, does not concern us here: see W. TREADGOLD, The Byzantine Revival 780–842. Stanford 1988, 374–375 and ALPERS 1988: 345–346. 100 Part One: Texts and Contexts happiness can only be achieved by tranquillity and peace of mind. He has no need of riches, fame or passions, but hopes to gain the magical plant, m0ly, that wards off evil thoughts. If only he could live up to these convictions of his till the day he dies! The poem is crammed with allusions to the Odyssey, referring not only to the mysterious “moly”, but also to the lotus-eaters, the gloomy cave of Circe and the enticing siren song. AP XV, 13 and 14 are two fiercely combative poems by Constantine the Sicilian and Theophanes the Grammarian. In the first poem Constantine brags about the professorial chair he holds. He proudly informs us that it is a seat of knowledge on which only highly educated people, like himself, are allowed to sit. His puffery is criticized by Theophanes in the next poem. “This chair of yours is no big deal. It is not of gold, not of silver, not of ivory. It is just a piece of wood. So, what are you bragging about? Anyone, scholar or fool, can sit on a wooden chair”. In the Anthologia Barberina, an early tenth-century collection of anacreontics and alphabets (see below, pp. 123–128), we find the same three names, Leo the Philosopher, Constantine the Sicilian and Theophanes the Grammarian, side by side in a section devoted to the anacreontics of ninth-century grammarians (nos. 58–64): Leo the Philosopher (58–59), Sergios and Leontios the Grammar- ians (60–61), Constantine the Grammarian [=Const. the Sicilian] (62–63) and Theophanes the Grammarian (64). Sergios and Leontios are mere names to us. Seeing that the title of Leontios’ anacreontic (no longer extant in the manuscript) clearly indicates that Leontios imitated an epigram of Agathias (AP V, 237)49, there can be little doubt that the Cycle of Agathias was already known to the circle of Leo the Philosopher. In fact, it will become abundantly clear that Leo the Philosopher and his students not only read, but also edited ancient epigrams several decades before Cephalas compiled his anthology. In a recent article I pointed out that one of the major sources Cephalas used for his anthology was the Palladas Sylloge50. This sylloge contained a lot of Palladas, of course, but also a number of epigrams or epic fragments by Lucian, Nestor of Laranda, Julian the Apostate, Cyrus of Panopolis, Claudian and many others. The sylloge was put together in the sixth century, probably between 551 and 567, in response to the fashionable revival of the epigram that was to lead to Agathias’ compilation of the Cycle. However, Cephalas did not have direct access to an original sixth-century manuscript, but made use of a ninth-century copy made by or for Leo the Philosopher51. Leo the Philoso- pher’s manuscript of the Palladas Sylloge also included a number of epigrams he had written himself: AP IX, 200–203, 214 and 578. These epigrams were

49 See LAUXTERMANN 1999a: 166–167 and CRIMI 2001: 39–40. 50 See LAUXTERMANN 1997. 51 See WIFSTRAND 1933: 169–170 and LAUXTERMANN 1999a: 161–163. Anthologies and Anthologists 101 copied by Cephalas along with the rest of the Palladas Sylloge. Cephalas incor- rectly ascribed to Leo the Philosopher two late antique poems, a cento and an epic fragment, because he found them next to authentic poems by Leo and erroneously assumed that they had been written by the same author. The cento consists of Homeric tags. It is a girl’s complaint about the painful experience of her defloration (AP IX, 361). The scabrous subject of this epigram is without parallel in Byzantine poetry, for if the theme is touched upon at all (for instance, in the Maximo scene in the Digenes Akrites), it is always viewed from the angle of male superiority, not from the perspective of the girl. Furthermore, all the other centos in the Greek Anthology date from late antiquity52, and there is no evidence that Byzantine poets, apart from the enigmatic author of the Christus Patiens, dabbled in the art of cento-writing. True, there are some Byzantine poems that have a lot of Homeric reminiscences, such as AP XV, 12 (Leo the Philosopher), 28 (Anastasios Quaestor) and 40 (Kometas), but none of these poems are real centos. The second poem incor- rectly ascribed to Leo the Philosopher, AP IX, 579, deals with Arethousa, the famous Sicilian water nymph. It is a fragment of a late antique mythological epic. As fragments rarely make sense, the poem is almost incomprehensible in its present form53. The Palladas Sylloge contained many epic fragments of this kind, such as, for instance, some passages from the Metamorphoses of Nestor of Laranda, all of which deal with aquatic subjects: rivers, sources, and so on54. The epic fragment on Arethousa might equally derive from the Metamor- phoses55, but even if it does not, it can safely be dated to the period of late antiquity and, therefore, cannot have been written by Leo the Philosopher. These two false ascriptions leave no doubt that Cephalas read the Palladas Sylloge in an updated version of the mid-ninth century composed by Leo the Philosopher himself or copied at his behest. There are more shreds and pieces of evidence indicating that Leo the Philosopher was familiar with ancient epigrams and played a significant role in the text history of the Greek Anthol- ogy. In a satirical poem on a stuttering student56 he coins the word

52 AP IX, 381–382 and Appendix Barberino-Vaticana no. 7 (ed. CAMERON 1993: 172). See also HUNGER 1978: II, 98–100. 53 See WESTERINK 1986: 195–196. 54 See the prooemium to the Metamorphoses (AP IX, 364); see also AP IX, 128–129 and 537. 55 AP IX, 536, which is probably a fragment of the Metamorphoses, also deals with the well- known story of the river Alpheios who, desperately in love with Arethousa, glides under the surface of the Adriatic to turn up again in Sicily. AP IX, 362, another epic fragment, treats the same subject, but does not belong to the Metamorphoses as its hexameters are post-Nonnian (see WIFSTRAND 1933: 168). 56 Ed. WESTERINK 1986: 200–201 (no. XI). 102 Part One: Texts and Contexts traylep5trayloß, which is formed by analogy with the neologism óaylep5óayloß found in AP XI, 238. The early tenth-century Sylloge Euphemi- ana (see pp. 114–115) contains a poem by Leo, in which he derides his doctor for prescribing a regime of cold water in the middle of winter57. The insertion of Leo’s poem in a collection of ancient epigrams indicates, I think, that its redactor wished to pay tribute to Leo the Philosopher for his scholarly work on the Greek Anthology. Finally, the fact that two of his students, Constantine and Theophanes, published collections of erotic epigrams, strongly suggests that the Greek Anthology was one of the many scholarly pursuits to which Leo the Philosopher turned his attention. The so-called Sylloge Parisina is divided into two parts deriving from two different sources. The first part contains a selection of epigrams from Cephalas’ anthology. The second part is a collection of pederastic epigrams headed by Constantine the Sicilian’s Love Song (îŸd1rion ™rztikön)58. This collection of pederastic epigrams is closely related to AP XII, one of the books of Cephalas’ anthology. But since the collection contains many pederastic epigrams that cannot be found in AP XII, it appears to derive from a source other than Cephalas’ anthology59. This source I call PCP (Parisian Collection of Paederas- tica). The main difference between Cephalas and PCP is that the latter does not confuse gender, whereas Cephalas had some trouble distinguishing boys from girls and regularly misclassified erotic epigrams. Take for instance AP XI, 51 and 53, which Cephalas mistakenly placed among the gnomic epigrams because he failed to understand their elusive meaning. The redactor of PCP, however, had no problem in grasping the sexual innuendo of these two epigrams and rightly recognized their pederastic nature. To give another example, Cephalas placed the famous epigram on Agathon by Ps. Plato in the heterosexual section: “I stayed my soul on my lips kissing Agathon. The rascal had come to cross over to him” (AP V, 78). This is truly a stupendous blunder. The redactor of PCP, once again, rightly judged that what we have here is one male in love with another. Given the fact that PCP contains epigrams not found in AP XII and does not present the sort of misclassifications typical of Cephalas, there can be but little doubt that it does not derive from the anthology of Cephalas. The original PCP is beyond any secure reconstruction, because the second part of the Sylloge Parisina appears to contain only a few excerpts. However, as the

57 Ed. WESTERINK 1986: 200 (no. X). 58 For a thorough description of the Sylloge Parisina, see CAMERON 1993: 217–245. The sylloge can be found in Par. Suppl. gr. 352 and Par. gr. 1630. For a description of these two manuscripts, see Appendix I, pp. 287–293, esp. pp. 291–292 and n. 21. 59 CAMERON 1993: 224 and 238–253, on the contrary, argues that the epigrams lacking in AP XII but found in the Sylloge Parisina ultimately derive from the anthology of Cephalas. But see LAUXTERMANN 1999a: 163–164, for a refutation of Cameron’s views. Anthologies and Anthologists 103 second part of the Sylloge Parisina and AP XII often have the same epigrams in the same order60, it would seem that PCP was one of the many sources used by Cephalas for the compilation of his anthology. The redactor of PCP can doubtless be identified with Constantine the Sicilian since the pederastic epigrams in the Sylloge Parisina start with his delightful poem on Eros, the Love Song in anacreontics61. Constantine the Sicilian wrote the poem ™n neöthti pa5fzn, oÊti spoyd1fzn, as the lemma attached to it states. Born in c. 825–83062, Constantine will have written the poem when he was still a student at the Magnaura school or shortly afterwards; but he may have added it to PCP in a later stage. For obvious reasons PCP must have been compiled before c. 870, when Constantine suffered his crise de conscience and publicly disavowed his former teacher, Leo the Philosopher. In the Love Song Constantine describes an unfortunate encounter with Eros: one day he catches sight of him, chases him in vain, and is then struck “below the waist” by the arrows of the little devil. In need of moral support the poet begs the chorus of his companions to join in the singing: “My friend, spend sleepless nights like Achilles singing in sweet harmony with the warbling nightingales. I have experienced the charms of love, but I do not find anywhere the way out. Give me a companion along the paths of song, to sing with me of Eros”. Since all the epigrams in PCP can be said to celebrate the power of Eros, PCP is in a sense the fulfillment of Constantine’s appeal to his fellow poets “to sing with him of Eros”. Thus the ancient epigrammatists and Constantine meet in the timeless space of intertext, where poetry is a substitute for real life and a compensation for the sorrows of love. Love may be unattainable, but one may “spend sleepless nights” with one’s friends and confess to them one’s deepest desires. Constantine the Sicilian’s Love Song is an appropriate introduction to PCP, for it shapes a fictitious setting of unrequited love and male bonding, and thus provides a context in which homo-erotic poetry may be read, interpreted and relished. Though Constantine wrote the poem when he was still a young man, he shows a remarkable erudition for someone his age. The poem abounds with all sorts of literary reminiscences: Moschus’ Runaway Love, Longus’ Daph- nis and Chloë and ancient epithalamia63. The borrowings from Moschus’ delight-

60 See CAMERON 1993: 242. 61 Ed. CRAMER 1841: 380–383 and MATRANGA 1850: 693–696. The lemma attached to the poem in Par. Suppl. gr. 352 does not mention the author. The index of Barb. gr. 310 preserves the original title: to¯ aJto¯ (i.e. Kznstant5noy grammatiko¯) îŸd1rion ™rztikñn di\ änakr6[ontoß], Ýper ðÍsen ™n neöthti pa5fzn, oÊti spoyd1fzn, Çlaben dê tën Üpöqesin ™k melùd5aß tinñß ãädom6nhß ™n g1mù: see GALLAVOTTI 1987: 39 and 49–51 and NISSEN 1940: 66– 67. 62 See LAUXTERMANN 1999a: 170, n. 27. 63 See R.C. MCCAIL, Byz 58 (1988) 112–122, CAMERON 1993: 249–252 and CRIMI 2001: 40–43. 104 Part One: Texts and Contexts ful Runaway Love are particularly interesting because the poem was included in the Garland of Meleager. This strongly suggests that Constantine the Sicilian was familiar with the contents of the Garland already at a young age, which may serve as an argument in favour of an early date for the compilation of PCP: say, in the late 840s or the 850s. Erotic epigrams and anacreontics seem to have been popular in the circle of Leo the Philosopher. Theophanes the Grammarian is the author of an anacreontic entitled in the index of Barb. gr. 310: “how he loves his friend and how he is not loved in return because of his extreme affection”64. Unfortunately, the anacreontic is not preserved in the manuscript, so we can only guess how Theophanes may have treated this daring theme without getting himself into trouble. Theophanes also wrote the following erotic epigram: “If only I could be a white lily so that you may put me close to your nostrils and satiate me still more with your skin” (AP XV, 35). The epigram is an obvious imitation of AP V, 83 and 84, the second of which reads in translation: “If only I could be a pink rose so that you may take me in your hand and put me between your snowy breasts”65. In the Palatine manuscript Theophanes’ epigram can be found near the end, but originally, in the anthology of Cephalas, it immediately followed AP V, 83–8466. The beginning of AP V (nos. 2–103) contains a great number of epigrams deriving from the so-called Sylloge Rufiniana. This was a small sylloge of erotic epigrams by the first-century poet Rufinus; since the sylloge also contained a few “Diogenianian” authors, such as Gaetulicus, Cil- lactor and Nicarchus, it was probably compiled by the second-century anthologist Diogenian67. It is impossible to reconstruct the original Sylloge Rufiniana, but we can identify in AP V at least three sequences of epigrams deriving from it (with additional material from other sources): AP V, 14–22, 27–51 and 66–84. Theophanes’ epigram and the two epigrams that he imitated are found at the end of the last sequence. What exactly has Theophanes to do with the Sylloge Rufiniana? Not an easy question, but we should bear in mind the overall design of Cephalas’ anthology. His anthology is basically a collection of Hellenistic, early Roman and late antique epigrams. That is why AP V–VII and IX–XIV do not contain contemporary poetry, with the tantalizing exception of a few

64 See the index in GALLAVOTTI 1987: no. 64. 65 For the text of these three epigrams, see CAMERON 1993: 283–285. Cameron supposes that AP V, 84 is also the work of Theophanes, but attaches too much importance to an incorrect reading in Arethas. He does not pay attention to the vocabulary. In AP XV, 35 Theophanes uses two Byzantine neologisms, croti8 (see ThGL) and ärg6nnaoß; AP V, 84, on the contrary, is written in Hellenistic Greek. 66 See CAMERON 1993: 283–285. 67 See P. SAKOLOWSKI, De Anthologia Palatina quaestiones. Leipzig 1893, 64–71, and CAMERON 1993: 84–90. Anthologies and Anthologists 105 epigrams by Leo the Philosopher and Theophanes the Grammarian. Leo’s epigrams are there because Cephalas used a ninth-century manuscript of the Palladas Sylloge copied by or for Leo the Philosopher. It is reasonable to conjecture that Cephalas included Theophanes’ epigram for exactly the same reason: Cephalas made use of a copy of the Sylloge Rufiniana made in the mid- ninth century by Theophanes and faithfully transcribed the epigram Theophanes had written himself at the end of the sylloge. Leo the Philosopher and his pupils evidently liked poetry, but while they were busy studying and copying epigrams, voices of dissent could be heard protesting against the mythological oddities and gross obscenities of ancient poetry. The entry on Theognis in the Epitome of Hesychius (c. 840–850) provides a good example: “Theognis also wrote gnomic epigrams, but among these you may find disgusting love poems on boys and many other things that are repugnant to those who live a pious life”68. Photios is another dissenting voice. In general Photios does not have much to say on the topic of ancient poetry, but its conspicuous absence in the Bibliotheca strongly suggests that he had little taste for the poets. In the entry on Empress Eudokia’s religious centos, however, Photios treats her with lavish deference and compliments her for telling the plain truth and not seducing the minds of young people with sweet lies69. Truth is beauty, but beauty is not necessarily truth. Photios objects to ancient poetry because of its contents, false and full of illusions, acting counter to the incontestable truths of Christianity70. In his view, classical poetry was at best only of secondary importance; it might provide students with the tools for acquiring a good style, but its role in the educational programme, as he envisaged it, had perforce to be ancillary71. This viewpoint is radically different from that of Leo the Philosopher. Leo and Photios are the greatest scholars of the ninth century, but apart from their immense erudition they really have

68 See M.L. WEST, Studies in Greek Elegy and Iambus. Berlin–New York 1974, 44–45. The Epitome of Hesychius is probably the work of Ignatios the Deacon: see W. TREADGOLD, The Nature of the Bibliotheca of Photius. Washington 1980, 31–32 and 36, and MANGO 1997: 4–5. 69 Photius. Bibliothèque. Tome II, ed. R. HENRY. Paris 1960, 195–196 (cod. 183). See B. BALDWIN, BMGS 4 (1978) 9–14 (Studies on Late Roman and Byzantine History, Liter- ature and Language. Amsterdam 1984, 397–402) and Aevum 60 (1986) 218–222 (Roman and Byzantine Papers. Amsterdam 1989, 334–338). 70 See, for instance, letters 56 and 209 (Photii patriarchae Constantinopolitani Epistulae et Amphilochia, ed. B. LAOURDAS & L.G. WESTERINK. Leipzig 1983–85, I, 103 and II, 109). See also ALPERS 1988: 357, n. 89, and 359–360. 71 See the comment by A. HEISENBERG, Historische Zeitschrift 133 (1926) 398: “Photios war weit davon entfernt eine klassische Philologie begründen zu wollen oder sich gar als Humanist zu fühlen”. See also H. HUNGER, Reich der Neuen Mitte. Graz 1965, 361. 106 Part One: Texts and Contexts nothing in common72. It is not difficult to guess, therefore, who of the two is the author of AP IX, 203 bearing the following lemma: Uzt5oy, oW dê L6ontoß. It is a laudatory epigram on Achilles Tatius’ novel Clitophon and Leucippe. The story is very decent, so we are told, not at all improper to read, for in the end the two heroes are rewarded for their chastity with the pleasures of blessed marriage. Since the novel is criticized in the Bibliotheca for its utter immorality, it is out of the question that Photios could have written this epigram73. Leo is a very likely candidate, not only because the erotic muse was much in vogue in the circle of Leo the Philosopher, but also because the Love Song by Constan- tine the Sicilian, one of his students, alludes to another ancient novel, Longus’ Daphnis and Chloë. Leo the Philosopher’s enthusiasm for classical literature was certainly not shared by all of his contemporaries, but as long as the great man lived, he dominated the intellectual scene of Constantinople with his presence. But when he died, the petty Telchines eagerly grabbed the chance to make a clean sweep, and sweep they did. After c. 870 there are no erotic epigrams and anacreontics, and though classicism is still much in vogue, no one any longer dares to study the ancients on their own terms without making excuses for it to orthodox fundamentalists. Cephalas feels obliged to put a statement of faith at the beginning of his anthology and begins his collection of Christian epigrams with a verse inscription inspired by Patriarch Photios. The name Pho- tios also pops up in connection with Constantine the Sicilian’s “conversion” to orthodoxy. The Psogos and the Apology are followed by a third poem74, in which Constantine claims to have discovered the source of salvation, albeit as an old man: now at last he knows that it is the Christian rhetoric of Photios that paves the way to heaven! The conflict between hellenism and orthodoxy also expresses itself in an unexpected source: the palindromes of the Greek Anthology. In the Planudean Anthology, but also in many other collections of ancient epigrams, we find a group of twelve palindromes: APl 387, nos. 1–4 and

72 J. HERGENRÖTHER, Photios, Patriarch von Konstantinopel. Sein Leben, seine Schriften und das griechische Schisma. Regensburg 1867, I, 323: “es scheint die Geistesrichtung beider Männer weit auseinandergegangen zu sein”; cf. Photios’ letter to Leo the Philos- opher: no. 208 (Laourdas & Westerink, II, pp. 107–108). See the excellent study by ALPERS 1988: 353–354 and 356–357. 73 See H.-G. BECK, Byzantinisches Erotikon. Munich 1984, 110–115. On Photios and the ancient novels, see P. AGAPITOS, in: Studies in Heliodorus, ed. R. HUNTER. Cambridge 1998, 128–132. 74 Ed. SPADARO 1971: 202. In the ms. the fourth verse reads: Ýß me g1lakti Çqreve qe5zn nam1tzn. WESTERINK 1986: 201 proposes the following excellent emendation: Ýß me g1lakti Š0n qr6ve qe5zn nam1tzn and suggests in the apparatus criticus to read Ös5zn instead of qe5zn. Anthologies and Anthologists 107

10; two verses not copied by Planudes; APl 387, nos. 6, 5, 8, 7 and 975. Palindromes are totally nonsensical, of course, but may betray a certain men- tality. Two of the palindromes deal with Photios and Leo the Philosopher, respectively: soò t/ óztò soUñß Étz ówtioß and n/ ™lat2 më ¸nht2 soUñß 4th no8mata l6zn76. The texts can hardly be translated but mean something like: “Let the wise Photios come to you with his light” and “Useless thoughts forged by the mind are baneful, wise Leo”. By putting the word soóöß right in the middle, the author of the two palindromes makes clear that he is opposing two types of wisdom, religious and profane. There can be little doubt that the author sides with the camp of Photios. Photios is the light shining forth, Photios is the intellectual guide leading the way. Conversely, the profane wisdom of Leo the Philosopher is useless, if not downright pernicious. Cephalas, Constantine the Sicilian and the anonymous author of the two palindromes pay lip service to the ideas of Photios. After c. 870 the Greek Anthology continues to be studied, but with the death of Leo the Philosopher dies the ideal of an enlightened hellenism. From that moment on, the legacy of hellenism has to be christianized in order to become acceptable.

** *

A Collection of Classicistic Epigrams: AP XV, 28–40

The small collection of epigrams copied at the end of the B manuscript, AP XV, 28–40, illustrates the ideological turnover from Leo to Photios, from unreserved enthusiasm for the ancients to a sort of classicism in Christian disguise. The collection was unquestionably compiled before 902, because the lemma attached to AP XV, 32 leaves no doubt that Arethas had not yet become archbishop when the collection was made. It cannot be excluded that the small sylloge was already to be found in the original Cephalas, but I am inclined to think that the epigrams were added to the anthology of Cephalas in what was undoubtedly one of its earliest apographs. The reason is the duplication of Theophanes’ epigram at AP XV, 35. If the sylloge had been put

75 See GALLAVOTTI 1989: 52–59 and 62–65, and STERNBACH 1900: 298–301. GALLAVOTTI 1989: 56–57 and 64 thinks that the initial collection consisted of 19 palindromes; however, since nos. 13–19 are not found in a fixed order in the manuscripts, I would suggest that they are later additions to the collection. 76 There can be no doubt about the identity of this L6zn, for the last eight palindromes including the one on L6zn are entitled in the various mss. containing the collection: L6ontoß Uilosöóoy. 108 Part One: Texts and Contexts together by Cephalas, it would mean that he had copied the epigram twice: first immediately after AP V, 83–84 and then again at AP XV, 35. Duplica- tions of this kind are fairly normal in the anthology of Cephalas, but in practically all the instances of duplication the most likely explanation is that Cephalas found the epigram in two different copies77. It is unlikely, however, that Cephalas found Theophanes’ epigram in two different copies of the Sylloge Rufiniana. For when an epigram is repeated, it is usually found in its original context: a Meleagrian author among other epigrams deriving from the Garland of Meleager, etc. However, in AP XV, 28–40 Theophanes’ epigram is “out of context”: it is no longer part of the Sylloge Rufiniana where it originally could be found, but figures among contemporary epigrams. So, if AP XV, 35 is not an ordinary instance of duplication, why did Cephalas copy it twice? And why did he copy it the first time with the correct reading ½is5n and then change it to cers5n?78 Regrettably, I cannot offer decisive proof, but I strongly suspect that AP XV, 28–40 was compiled by someone other than Cephalas. Due to a binding error the order of the epigrams in the sylloge has been reversed79. The original order is as follows: no. 40 and then nos. 28 to 39. The sylloge appears to have a thematic structure. It starts with two poems in pseudo-Homeric style by Kometas and Anastasios Quaestor (AP XV, 40 and 28). Then we have a number of epitaphs: AP XV, 29–31 by Ignatios the Deacon and AP XV, 32–34 by Arethas. This in its turn is followed by The- ophanes’ erotic epigram (AP XV, 35). The sylloge ends with six book epigrams: on an edition of Homer by Kometas (AP XV, 36–38), on a Homeric grammar by Ignatios the Deacon (AP XV, 39, v. 1 and vv. 2–3), and on Plato by an anonymous author (AP XV, 39, vv. 4–5). The author of AP XV, 40, Kometas, is not entirely unknown to us. He was appointed professor of grammar at the Magnaura school in the 840s and produced a punctuated edition of the two Homeric epics80. In AP XV, 36–38, epigrams that served as an introduction to this edition, Kometas emphasizes the magnitude of the problems he faced when he transliterated Homer from

77 See CAMERON 1993: 43–48. 78 See CAMERON 1993: 284. 79 After the binding error had been made, scribe J rewrote in the top margin of p. 693 the first nine lines of AP XV, 28, which had become acephalous. In the lemma he added the nickname of Anastasios Quaestor: Ö Traylöß, and in v. 2 he supplemented a lacuna: Çczn. There is no need to suppose that he used another manuscript to come up with these two insignificant additions. 80 Theoph. Cont. 192, 19–20. On Kometas and his scholarly work on Homer, see LEMERLE 1971: 166–167, R. BROWNING, Viator 6 (1975) 22–23 (repr. in: idem, Studies on Byzantine History, Literature and Education. London 1977, no. 17), ALPERS 1991: 254–257, and G. CORTASSA, Prometheus 23 (1997) 222–228. Anthologies and Anthologists 109 uncial to minuscule. Kometas might be slightly exaggerating, but as any modern editor knows, the problem of punctuation can indeed be troublesome, for it necessarily presupposes that one fully understands the text one is editing. Homeric Greek is not always easy to understand and Kometas is therefore likely to have made use of ancient commentaries or marginal scholia whenever he stumbled upon a difficult passage in Homer. Kometas’ edition is not preserved, but in AP XV, 40 he quotes five lines from Homer in full, among which Il. 2. 87 (v. 35) with the reading 3din1zn, “corrected” in all modern editions of the AP to ädin1zn, although the breathing was recommended by the great Homeric scholar Aristarchus81. However, there are serious reasons to question Kometas’ claim that he produced a reliable edition of Homer, for AP XV, 40 “is perhaps the single most unmetrical poem in the Anthology”. Kometas has no feel for the hexameter and commits really awful prosodic errors – “poor qualifications for a ‘restorer’ of the text of Homer”82. Against AP XV, 37 and 40 Constantine the Rhodian scribbled in the margin a few satirical verses criticizing Kometas for his lack of poetical skills83. The following is a good specimen of Constantine’s talent to abuse: “Kometas, you were another Ther- sites. So, how did you dare to impersonate Achilles, you wretch? To hell with these products of an unpoetical mind! Off to the gallows, off to the pillory with these verses full of the rottenness of dung!”. Constantine the Rhodian obviously objected to Kometas’ claim to be an expert in Homer given the poor quality of his hexameters. But the histrionic metaphor in the second verse (\Acill6zß pröszpon eœs6dyß) appears to indicate that there was yet another aspect to Kometas that Constantine found extremely offensive: his false pretence. In the view of Constantine, Kometas is putting on a mask in AP XV, 40. The poem simply lacks sincerity. To understand Constantine’s angry reaction, it suffices to take a closer look at AP XV, 40. It is a fifty-seven-line poem on the Raising of Lazarus. It paraphrases in Homeric Greek chapter 11 of the Gospel according to John; however, the author does not simply retell the biblical story, but expands on the theme84. Kometas needs many verses to express what the Bible says in a few words. For instance, he turns the simple sentence: k7rie, eœ Íß ¢de oJk 4n moy äp6qanen Ö ädelóöß (11: 21 and 32) into two highly elaborate verses: eœ g2r t!de

81 See the commentary ad locum in: The Iliad, ed. W. LEAF. London 1900–1902 (repr. Amsterdam 1971). 82 CAMERON 1993: 309. 83 Ed. BECKBY 1957–58: IV, 288 and 292 (at AP XV, 37 and 40). See F.M. PONTANI, in: Studi in onore di Aristide Colonna. Perugia 1982, 247–253, and CAMERON 1993: 309–310. 84 See M. CAPRARA, Koinzn5a 24 (2000) 245–260, who argues that Kometas was familiar with Nonnos’ Paraphrase of the Gospel according to John and owed his inspiration to this classic example of biblical paraphrase in verse. 110 Part One: Texts and Contexts

Çhsqa, 4nax nek1dzn \Aúdzneáß / oÊpot\ Çtlh me¦nai, ™peò Í polá ó6rtatoß Ísqa (vv. 43–44). The words of the two sisters of Lazarus, Maria and Martha, are highly emotional in the biblical version, but are devoid of any concrete meaning in Kometas’ poem, and thus the deeply felt sorrow of bereavement evaporates into thin air. This is in general Kometas’ problem: he keeps heaping up magnil- oquent words, but none of these words signify anything else than a painful dearth of feeling. His poem is simply a bad poem, the product of a frigid muse. However, since it is certainly not the only bad poem written in Byzantium, one may wonder why Constantine the Rhodian reacted as he did. I think that his reaction is one of sincere disappointment. The story of Lazarus is fundamental to Christianity, for it epitomizes one of the quintessential tenets of Christian faith, namely the resurrection of the dead. It is the prelude to the Anastasis of Christ. Death is defeated, eternal life is near at hand. With all its theological connotations, the Raising of Lazarus is a story of hope and happy expectations – a moment of intense joy relived each year on the last Saturday before Easter. By turning the story into a sterile exercise in the art of rhetoric, Kometas failed to convey the message of this liturgical feast to his Byzantine audience. The poem next in line is AP XV, 28 by Anastasios Quaestor, also known as the “Stammerer” (Ö traylöß)85. Anastasios was born in the later ninth century and died after 922; he was a close friend of Leo Choirosphaktes and an adver- sary of Arethas; he took part in the Doukas revolt (913), was imprisoned in the Stoudios monastery and regained his former position when Romanos Lekap- enos assumed power (919). Anastasios wrote an encomiastic epitaph on Metrophanes of Smyrna and a satirical poem on the death of Emperor Alexan- der. He is also the author of various iambic canons in the classicistic style of Ps. John of Damascus86. AP XV, 28 describes the scene of the Crucifixion: Christ on the cross with the two thieves on either side, the Virgin Mary and John the Apostle, some wayfarers who make fun of Christ, and “the people of the Jews” offering Him sour wine to drink. The poet depicts the whole scene with short, vivid brushstrokes, painting as it were in words, and guides our mind’s eye by presenting the participants and their reactions one after the other in a narra- tive sequence. In the first verses he uses descriptive imperfects, but when he portrays “the wicked and bloodthirsty people of the Jews”, he suddenly uses an aorist, Ërexe pot‰ta, and thus draws attention to their lewd action. He ends his description by saying that Christ, who is both Man and God, “was silent and resisted not”. The poem might well have ended here, but we find to our surprise three additional verses prescribing the appropriate viewer’s response to the scene: “Who would be so stupid as to be full of pride when he reflects on

85 On the tumultuous life of the author, see LAUXTERMANN 1998a: 401–405. 86 Ed. PAPADOPOULOS-KERAMEUS 1900: 43–59. Anthologies and Anthologists 111 this in his heart and sees it in pictures? For as God He prevails over us, but as Man He does not”87. The poem is strongly anti-Semitic, but by the sudden twist at the end it becomes clear that arrogant Christians are in no way better than the Jews who jeered at Christ. When the viewer looks at the awesome mystery of the Son of God dying on the cross, his attitude should be one of humility. It is not clear whether Anastasios had a particular picture in mind when he wrote the poem, but the word gymnöß indicates that he was thinking of contemporary representations of the Crucifixion, in which Christ was seen wearing a loincloth instead of the earlier colobium. Anastasios’ poem is full of Homeric reminiscences, but where the similar experiment by Kometas failed, Anastasios succeeds in getting his poetic message across. The hexameters are almost flawless except for one or two venial slips. Homer is not the only source of inspiration, for Anastasios uses the Sophoclean word l7gdhn (“in sobs”), the Hellenistic adjective dival6oß, the rare form kirn1menoß, the poetic Örwmenoß and the hapax aWmatoc1rmhß. The poem is all in all a splendid example of a Christian theme treated in a classicizing manner. AP XV, 29–31 are three epitaphs in elegiacs by Ignatios the Deacon, the well-known author of the first half of the ninth century88. In its detailed entry on Ignatios the Deacon and his various literary works, the Souda mentions the following category: ™pitymb5oyß ™l6goyß89. The three epitaphs preserved in the Palatine Anthology belong to this category, but there can be but little doubt that the category comprised more than the three specimens still extant. The Souda clearly refers to a collection of epitaphs – a collection now lost, but still available to the person who compiled AP XV, 28–4090. Ignatios may have conceived the idea of producing a collective edition of his epitaphs by analogy with the similar collection of Gregory of Nazianzos’ ™pit7mbia ™pigr1mmata. The latter seems to have been quite popular in the middle Byzantine period, given the number of early manuscripts containing sepulchral epigrams by Gregory of Nazianzos: the Palatine manuscript (twice: AP VIII and the collection copied by J on the last pages), Bodl. Clark. 12 (s. X), Laur. VII 10 (s. XI) and Ambros.

87 In the last verse I follow the interpretation of P.T. BRANNAN, American Journal of Philology 80 (1959) 396–399. 88 For the life and works of Ignatios, see WOLSKA-CONUS 1970: 330–351, MANGO 1997: 1–24, MAKRIS 1997: 3–22, LAUXTERMANN 1998a: 397–401, S. EFTHYMIADIS, The Life of the Patriarch Tarasios by Ignatios the Deacon. Introduction, Text, Translation and Com- mentary. Aldershot 1998, 38–46, KAZHDAN 1999: 343–348, and TH. PRATSCH, BMGS 24 (2000) 82–101. 89 Ed. ADLER 1928–1938: II, 607–608. 90 Perhaps the collection of epitaphs was headed by Ignatios’ funerary anacreontic (ed. CICCOLELLA 2000a: 42–54); cf. Constantine the Sicilian’s sylloge of pederastic epigrams (PCP), which also begins with an anacreontic. 112 Part One: Texts and Contexts gr. 433 (s. XI)91. What is more, Ignatios the Deacon did not hit upon the unusual idea of writing a sepulchral eis heauton (AP XV, 29) all by himself, but probably adopted the idea from Gregory of Nazianzos (cf. AP VIII, 80–84 and Greg. Naz. II, 1, 99). In AP XV, 29 Ignatios speaks to us from the grave, confesses his sins and prays to God for mercy92. AP XV, 30 and 31 are ordinary encomiastic epitaphs: the first praises a young man called Paul for the virtue and intellectual brilliance he displayed when he was still among the living; the second celebrates Samuel, a deacon of the Hagia Sophia, who showed his Christian zeal and piety by bequeathing his earthly possessions to the church. The language and style of these three epitaphs is obviously classicizing, but it is impossible to identify a particular literary model imitated by Ignatios: we find Homeric endings, such as -oio, -øsi and -essi, but Byzantine elegiacs in general make use of Homeric forms; Ámmati eJmen6i may be an imitation of Ámma eJmen6ß in AP VIII, 248. 2 by Gregory of Nazianzos, but late antique and Byzantine poetry is fond of the word Ámma (“the eye of Justice, the Emperor, God Almighty”, etc.)93; parallels for the rare expression ™n lagönessi aÉhß (“in the womb of earth”) can be found in ancient inscriptions94, but was Ignatios familiar with these parallels? Ignatios does not make prosodic errors, but rather surprisingly treats the caesura of the pentameter as a full stop where hiatus and even brevis in longo are allowed (29. 6; 30. 2 (!); 30. 4; 31. 2). The two book epigrams AP XV, 39, v. 1 and vv. 2–3 clearly indicate that Ignatios the Deacon did some scholarly work on Homer: see their title: “on the same”, that is, on Homer (the subject of AP XV, 36–38), and see the phrase soó‰ß pol7idriß äoid‰ß95. Ignatios proudly states that he “has brought to light the science of grammar hidden in the ocean of oblivion” – which is probably a gross exagger- ation, but at least gives a clue as to Ignatios’ precise contribution to the field of Homeric scholarship: grammatical epimerisms on Homer96. Since Homeric

91 See H.M. WERHAHN, Übersichtstabellen zur handschriftlichen Überlieferung der Gedich- te Gregors von Nazianz, in: W. HÖLLGER, Die handschriftliche Überlieferung der Gedich- te Gregors von Nazianz. 1. Die Gedichtgruppen XX und XI. Paderborn 1989, 34. 92 The last two verses were re-used by the scribe of Laur. LXX 20 (s. XI) as a token of his humility: ed. BANDINI 1763–70: II, 680 and COUGNY 1890: IV, no. 116. The epigram can also be found in Laur. XXXII 16 (see below, n. 119) and in Barb. gr. 74, Allatius’ collection of Byzantine poems (the source used by Allatius is the Palatine manuscript itself, which was in Rome at the time). 93 See, for instance, ROBERT 1948: 17, 25 and 138. 94 See CIG 7. 117 and 14. 2001. 95 In Ignatios’ letters no pagan author is quoted as often as Homer: see the Fontes in MANGO 1997. 96 On ninth-century Homeric epimerisms (by Choiroboskos?), see A.R. DYCK, Epimerismi Homerici. Pars prior epimerismos continens qui ad Iliadis librum A pertinent. Berlin– New York 1983, 5–7; Pars altera epimerismos continens qui ordine alphabetico traditi sunt. Berlin–New York 1995, 23–24. Anthologies and Anthologists 113 epimerisms already existed in late antiquity97, I suspect that Ignatios’ “redis- covery of grammar” entailed little more than producing a faithful copy of a late antique manuscript with some additional information gathered from other sources. The epitaphs by Arethas, AP XV, 32–34, are probably the worst poems ever written in ninth-century Byzantium. The poems on the death of his sister Anna, AP XV, 32–33, probably date from the 870s or the early 880s, seeing that she sadly died at the premature age of twenty-three. Unless we assume that he was much older, Arethas (born c. 850)98 will have been in his twenties or his early thirties when his sister died. The epitaph on the nun Febronia, AP XV, 34, may date from the same period as well. We happen to know a certain Febronia, born about 810, who founded a monastery and was renowned for her piety and erudition99. With all the erudition and poetic talents she is credited with, Febronia may have been capable of understanding and appreciating the tortuous style of Arethas, which is more than we can say for ourselves. Take for instance the second epitaph on the death of Anna written in dodecasyllables with harsh enjambments offending the ear (33. 3–4 and 9–10) and with ugly parentheses disrupting the natural flow of the verses (33. 2–3 and 7–8). It is impossible to recite the poem without faltering. A poem that cannot be heard is poetically dead – as dead as the sister whose passing-away Arethas bewails with many highfalutin words, but without ever convincing us that he truly mourns. The epitaph also lacks any reference to the spiritual salvation after death, for which the Byzantines longed so dearly. What are we to think of this? Did Anna not desire to be awarded a place in heaven? Did her family not care about her future in the hereafter? Of course they did. And so did Arethas, but he was more interested in words than in emotions. The epitaph on Febronia runs more smoothly than the two poems on Anna, but still lacks in stylistic dexterity. It begins as follows: “Febronia must surely have given some token of her sympathy to the spirits below likewise, if there, too, the poor have need of the wealthy”. The idea that the dead dwell in the limbo of Hades is common in Byzantium, of course, but no Byzantine believed that the poor even needed charity in the nether world. Neither did Arethas, but he simply used a classicistic oxymoron to emphasize Febronia’s virtue. In the next verses Arethas presents his own version of the Nekuia: “For not even there do the souls of the

97 On the Homeric epimerisms attributed to Herodianus, but dating from the sixth century, see DYCK, o.c., II, 37–40. 98 See KOUGEAS 1913: 1–9. 99 See I. VAN DEN GHEYN, AnBoll 18 (1899) 234–236. The hagiographer praises Febronia for her erudition: ta¦ß qe5aiß mel6taiß, Çti dê poihtik! kaò grammatik! kaò to¦ß t0n qe5zn pat6rzn ™mm6troiß pon8masin Çgnz Šaytën ™pido¯nai (234, 11–13), and he calls her ™llögimoß kaò pe¦ran œscyr2n ™n ta¦ß graóa¦ß Çcoysa (236, 6–7). 114 Part One: Texts and Contexts generous forget entirely their beneficence”. This is Homer all over again: the souls of the dead remembering their former life on earth. But then Christianity brutely intrudes into the Homeric scene: Febronia is compared to the biblical virgins who kept their oil lamp burning while waiting for the divine bride- groom. Febronia kept her oil and wicking alight by her charity to the poor. That is why she reposes in her tomb deeply asleep, but certain of entering the bridal chamber of Christ. Arethas has no feel for the elegiac: verses without caesura (32. 1, 3, 11; 34. 1 and 9), ugly sounding spondaics (e.g. 32. 5; 34. 5), neglect of bridges (32. 13; 34. 3, 5 and 9), etc. The classicistic sylloge of AP XV, 28–40 closes with an anonymous book epigram on a certain scholar who prepared an annotated edition of Plato or perhaps a commentary on the Platonic corpus (AP XV, 39, vv. 4–5). The poem probably dates from the late ninth century in the light of the fashionable revival of Plato at the time100. It is highly unfortunate that the B manuscript does not record the name of the author of the epigram, because the odds are that he was the same person who compiled the classicistic sylloge and who owned the exemplar copied by the B scribes. For, as we shall see below, owners of a manuscript of Cephalas’ anthology usually add epigrams of their own, thus allowing us to reconstruct the text history of the Greek Anthology.

** *

Constantine the Rhodian and Others

The anthology of Cephalas must have been a tremendous success right from the start given the great number of tenth-century manuscript copies; these are now all lost except for the Palatine manuscript, but there is ample evidence of them. The Sylloge Euphemiana contained various excerpts from the anthology of Cephalas rearranged in a new order. The original sylloge is lost, but we possess two independent sources that derive from it: a late fifteenth-century version of the sylloge (regrettably with substantial omissions) and the epigrams copied by the twelfth-century scribe Sp in the Palatine manuscript101. The Sylloge Euphemiana is named after the person to whom it is dedicated, Euphemios. Its author is unknown, but in the two dedicatory epigrams that accompany the sylloge, he informs us that he was born in Hypata in Thessaly

100 See J. IRIGOIN, Cahiers de Civilisation Médiévale 5 (1962) 287–292, LEMERLE 1971: 167– 169, and ALPERS 1991: 260–267. 101 See CAMERON 1993: 254–277. Anthologies and Anthologists 115

(Neai Patrai) and now resides in Constantinople where he loyally serves the Emperor Leo VI102. Fortunately, we know a little more about Euphemios, to whom the anonymous author dedicated “these few lilies from Helicon". There is a verse inscription from Attaleia commemorating the construction of a second fortification wall in 911–912 built by the mystographos Euphemios at the behest of the reigning emperors, Leo VI and Constantine VII103. There is also a satirical verse on Niketas Magistros quoted in the De Thematibus: gar- asdoeidëß Áviß ™sqlabzm6nh, “a Slavic face with a cunning look”104. Euphemios, “the famous grammarian” as he is called, wrote this verse to make fun of Niketas Magistros who boasted about his noble descent, although he was born in the Peloponnese, a backward province that had been overrun by Slavic tribes. It is reasonable to assume that the satirical poem, of which only this verse has been preserved, dates from 928 or shortly afterwards when Niketas had fallen into disfavour with the Lekapenos clan105. The Sylloge Euphemiana probably dates from the first decade of the tenth century: before 912 (the end of Leo’s reign) and after 890–900 (the date of Cephalas’ anthology). The sylloge contains three contemporary poems: the two dedicatory epigrams and a satirical poem by Leo the Philosopher directed against his nitwitted doctor. The Planudean Anthology derives its epigrams from two tenth-century sources, both of them abridged versions of the original anthology of Cephalas: Pla and Plb. The first source used by Planudes, Pla, contained a group of dodecasyllabic epigrams on famous charioteers of the past (APl 380–387); these were headed by an epitaph in elegiacs on the tenth-century charioteer Anasta- sios (APl 379)106. The epitaph was written by Thomas the Patrician and Log- othetes tou Dromou, a well-known figure in the history of early tenth-century Byzantium: Logothetes in 907 and 913, a correspondent of Leo Choirosphaktes and Arethas, a relative of the historian Genesios, and an intellectual renowned for his knowledge of philosophy107. It is reasonable to assume that Pla was a

102 Ed. COUGNY 1890: III, 256–257, WESTERINK 1986: 201, and CAMERON 1993: 255. 103 Ed. GRÉGOIRE 1922: no. 302. 104 Ed. PERTUSI 1952: II, 6, 33–42. See also P. SCHREINER, in: Festschrift H. Bräuer. Köln 1986, 487. 105 The precise date of De Thematibus is disputed, but I follow Kresten’s proposal for a date in the 960s (see I. ŠEVCENKO, in: Byzantine Diplomacy, ed. J. SHEPARD & S. FRANKLIN. Aldershot 1992, 185, n. 47): a dating supported by the word ™ke¦non in EJó8mion ™ke¦non tñn periböhton grammatikön, which indicates that the author of De Them. refers to the events of 928 as something of the past. 106 The epigrams can be found in Pla IV, 6, which forms an appendix to Pla IV, 3 (the late antique charioteer epigrams, nos. APl 335–378 and AP XV, 41–50). On APl 380–387, see chapter 5, pp. 173–179. 107 See A. MARKOPOULOS, ZRVI 24–25 (1986) 103–108 and CAMERON 1993: 319–320. 116 Part One: Texts and Contexts

Cephalas manuscript copied at the behest of Thomas himself or one of his friends. The second source used by Planudes, Plb, is connected with the name of one of the most prominent scholars of tenth-century Byzantium, Alexander of Nicaea108. Plb contained three epigrams written by Alexander: a witty epigram on a bath in Prainetos (APl 281)109 and two epitaphs to Nicholas Mystikos (APl 21–22)110. Again, it is very likely that Plb was copied by or for Alexander of Nicaea. Pla and Plb derive from two early tenth-century manuscripts containing the anthology of Cephalas plus a few contemporary epigrams added by their rightful owners. The Palatine Anthology, too, contains a collection of epigrams put together by the very person who had commissioned the manuscript and did the final editing, Constantine the Rhodian (scribe J). Constantine the Rhodian was born at Lindos in c. 880111. His well-informed marginal scholia on Gregory of Kamp- sa and Cephalas, which tell us who did what, clearly indicate that he knew these scholars personally, and suggest by implication that he was a student at the school of the New Church in the 890s112. Constantine definitely had a talent for verbal abuse, as borne out by the great number of satirical poems that go under his name. It is hardly surprising, therefore, that the powerful court eunuch Samonas availed himself of Constantine’s obliging services, made him his personal secretary and ordered him in 908 to write a libel against a favour- ite of Leo VI113. In the years 913–920 Constantine wrote the first version of the Ekphrasis of the Church of the Holy Apostles for the entertainment and instruction of the young prince Constantine VII. In 927 he went on an embassy to the Bulgarians to negotiate peace, by which time he had obtained a post in the palace clergy as basilikñß klhriköß114. Between 931 and 944 he wrote the second, enlarged version of the Ekphrasis, in which he praises the Lekapenoi115. Shortly after 944 he produced the Palatine Anthology. The date of his death is unknown. On pp. 666–668, between John of Gaza’s Ekphrasis and the Technopaegnia, Constantine the Rhodian copied as many epigrams as the available space permitted; and on pp. 670 and 673, below the Technopaegnia, he copied a few

108 On this scholar, see MARKOPOULOS 1994c: 313–326. 109 See P. MAAS, BNJ 3 (1922) 333–336 (repr. in: idem, Kleine Schriften. Munich 1973, 468– 472). 110 See CAMERON 1993: 317–319. See also ŠEVCENKO 1987: 462. 111 For the life of Constantine the Rhodian, see DOWNEY 1955: 212–221. 112 See CAMERON 1993: 108–116. 113 Theoph. Cont. 376, 1–4. See R. JENKINS, Speculum 23 (1948) 234 (repr. in: idem, Studies on Byzantine History of the 9th and 10th Centuries. London 1970, no. 10). 114 Theoph. Cont. 413, 1–3. 115 On the two versions of the Ekphrasis, see SPECK 1991: 249–268. Anthologies and Anthologists 117 more. Since AP XV, 1 belongs to the preceding Ekphrasis, the collection of epigrams compiled by Constantine the Rhodian begins only at AP XV, 2. The collection comprises the following epigrams: AP XV, 2–17; I, 122; IX, 400 and 180–181; XV, 18–19; X, 87; XV, 20; X, 95; XV, 23; and IX, 196–197. As the AP numbers already indicate, the collection contains a great number of duplications: epigrams that can also be found elsewhere in the Palatine manuscript. The reason for this is that Cephalas (for his anthology) and Constantine (for his collection) made use of the same source: the Palladas Sylloge. The Palladas Sylloge is not only the source for these doublets, but also for epigrams XV, 9– 10, 18–20 and 23. XV, 2–8 and 11, on the contrary, are verse inscriptions copied in situ by or for Constantine the Rhodian, and XV, 12–17 and I, 122 are Byzantine poems116. XV, 12–14 and I, 122 are poems by ninth-century intellectuals who contributed to the Greek Anthology: Leo the Philosopher, Constan- tine the Sicilian, Theophanes the Grammarian and Michael Chartophylax117. And XV, 15–17 are epigrams by Constantine the Rhodian himself, which he added to the manuscript because it was his own personal copy of Cephalas’ anthology. The manuscript was his, not only in terms of legal ownership, but also because he actually contributed to the copying and did the final editing. This is also why Constantine, like so many other Byzantine scribes, used the epithet tapeinöß in the lemma attached to AP XV, 15 as a means of signing his own work without appearing too vainglorious118. Incidentally, I would like to point out that it is wrong to call AP XV a “book” and to treat it as if it were a homogeneous compilation of epigrams. In fact, this particular book is an invention of modern editors who bracketed together all the epigrams and short poems they found at the end of the Palatine manuscript with an utter disregard of palaeographical and codicological data. This so-called book was written by three different scribes: XV, 28–40 by B3 (c. 920–930), XV, 1–27 by J (Constantine the Rhodian, shortly after 944), and XV, 41–51 by Sp (twelfth century). The first part (XV, 1–27) and the second part (XV, 28–51) of this phantasmal book are divided by a quaternion containing the Anacreontea. Furthermore, the Hellenistic Technopaegnia (XV, 21–22 and 24–27) are not epigrams and were certainly not intended by Constantine the Rhodian to be viewed as such. At the tail end of his own manuscript Constantine put poems that were of interest to him: John of Gaza’s Ekphrasis,

116 See LAUXTERMANN 1997: 329–330 and 334–335. 117 Treated above on pp. 99–100. CAMERON 1993: 307 asserts that the word mak1rioß in the lemmata attached to XV, 13–14 indicates that Constantine and Theophanes “had only recently died”. But the word simply indicates that they are dead and that the lemmatist feels respect for them. See, for instance, Ambr. E 100 Sup. (s. XIII), fol. 135: to¯ makar5oy \Iz1nnoy to¯ Gezm6troy, a lemma written some 200 years after the death of Geometres. 118 See CAMERON 1993: 304. 118 Part One: Texts and Contexts the Technopaegnia, the Anacreontea, a number of poems by Gregory of Nazian- zos as well as the collection of epigrams I just mentioned. Constantine the Rhodian had nothing to do with AP XV, 28–40 (copied some twenty years earlier by scribe B3) or with AP XV, 41–51 (copied some two centuries later by scribe Sp). There is no book AP XV. It is to be hoped that future editors will take this into consideration and future scholars will stop referring to AP XV as a separate book119.

** *

Byzantine Classicism and Modernism

Although the anthology of Cephalas was widely read in tenth-century Byzantium, and probably also in later periods120, it had barely any impact on Byzantine poets and did not significantly influence the course of Byzantine poetry. Only few Byzantine poems display the epigrammatic concinnity, the sense of poetic closure, the elegant technique of the elegiac and all the other fine qualities that make the classical epigram what it is: grand poetry in miniature. Only rarely does one stumble upon obvious literary reminiscences and only rarely can one identify an ancient epigram as the direct literary model for a Byzantine poem. Cameron pointed out that Geometres borrowed the word 4eisma (Cr. 281, 17) from Gregory of Nazianzos (AP VIII, 9. 1 and 113. 1), imitated a pythian oracle in Cr. 281, 14, and adapted an epigram by Palladas (AP XI, 386) in Cr. 331, 6121. To this list of literary reminiscences one may add the following poems. The elegiac poem, no. Cr. 340, 25, which deals with the unlucky fate of a fawn that was hunted down, jumped into the sea and died there in fishing nets, obviously imitates an epigram by Tiberius Illustris (AP IX, 370). The satirical poem on a eunuch, Cr. 293, 2, imitates a well-known

119 Laur. XXXII 16 (a. 1280–83) contains two collections of epigrams: on fols. 3–6 and 381– 384 (see CAMERON 1993: 201–216). The first collection ultimately derives from the Pala- tine manuscript, as shown by the following series of epigrams: AP XV, 9; epitaph to the wife of emperor Maurice; AP XV, 29. XV, 9 was added to Cephalas’ anthology by scribe J; XV, 29 was copied by scribe B. The surprising combination of XV, 9 and 29 in Laur. XXXII 16 points in the direction of the Palatine manuscript as the most likely source. 120 For the text history of the Greek Anthology in the twelfth century, see CAMERON 1993: 128–129 and 340–341. It should also be borne in mind that Planudes and other Palae- ologan scholars may well have found their tenth-century sources in manuscripts of the eleventh or twelfth century. 121 See CAMERON 1993: 337–339. Geometres adapted Palladas’ epigram also in his Dispute: ed. GRAUX 1880: 278 (no. 6). Anthologies and Anthologists 119 epitaph to Homer: AP VII, 3. At Cr. 320, 14 Geometres quotes Menander, Monostich 231. Geometres’ two poems on Summer (Cr. 316, 3 and 316, 11) borrow their imagery from epigrams on the beauty of nature (for instance, AP X, 1 and VIII, 129), and his long Ekphrasis of Spring (Cr. 348, 16) has much in common with a fourth-century ecphrastic poem by a certain author called Meleager (AP IX, 363)122. Taking into account the sheer bulk of Geometres’ poems, this list of reminiscences is hardly impressive. Sure enough, if one continues the search for parallels, the poetry of Geometres may provide more instances of literary imitation, but for every poem that is vaguely classicizing, there are dozens of poems that are certainly not. It is beyond doubt that Geometres was familiar with the anthology of Cephalas, but he had little taste for it, and the kind of poetry he wrote had little in common with ancient epigrams. The same is true for later Byzantine poetry in general: except for the occasional literary borrowing, there is no proof that it was influenced or even slightly affected by the ancient epigram. Most Byzantine epigrams do not classicize; they “modernize” (“modern” meaning anything written after c. 600, that is, “modern from a Byzantine perspective”). The ancient epigram exercised a strong influence over Byzantine poets only in the hundred years of classicism that began with Leo the Philosopher and ended with the compilation of the Palatine Anthology. Before c. 850 and after c. 950 ancient epigrammatic poetry has no place in the literary universe of the Byzantines; they may have read and even liked classical epigrams, but they did not feel the urge to imitate. However, in the hundred-year interval of c. 850–950 classicism is much in vogue. In the sections above, I treated this classicizing vogue in much detail and presented abundant evidence for it, so there is no need to discuss it again. It is perhaps worth noticing, however, that the classicizing vogue does not express itself only in literary epigrams, but also in verse inscriptions. The first example is the famous inscription in Skripou (the ancient Orchomenos), which dates from 873–874123. The poem is written in almost impeccable hexameters124 and its language is profoundly Homeric. See, for instance, ƒ pol7aine L6on formed by analogy with ƒ pol7ain\ \Odyse¯ (Il. 9. 673); the Homeric construction: participle + per Çmphß (=ka5per + participle); postponed ™pe5 in Çrga ™pe5… The Holy Virgin is called œói1nassa (!), probably by analogy with her cult title pant1nassa, but also as a learned allusion to Od.

122 For a comparison of these two poems on Spring, see KAMBYLIS 1994–95: 33–40. For the date of the poem by Meleager (not the famous poet and anthologist of the first century BC!), see WIFSTRAND 1933: 168–170 and CAMERON 1982: 231–232. 123 Ed. OIKONOMIDES 1994: 483–484. Read polycand6ú (v. 2), ™xet6lessaß (v. 4) and ×staton (v. 7). 124 But see the hiatus in v. 3: Çrga ™pe5, and the epic lengthening of a short vowel in v. 7: Cristo¯ d\ Škat6rzqen. 120 Part One: Texts and Contexts

11. 284: (king Amphion) Ýß pot\ ™n \Orcomen/ Minye5ù Éói 4nassen [cf. v. 12: (Leo) c0ron ™pikrat6zn te palaió1toy \Orcomeno¦o]. The poet was also familiar with the Greek Anthology: qeod6gmzn, a hapax recorded in AP VII, 363. 4; kaò oJ lal6onta, cf. APl 30. 4; ™p\ äpe5rona k7kla, cf. AP IX, 468. 3; mhtrñß äpeirog1moy, cf. AP I, 2. 3, 27. 3 and 99. 6; ™xet6lessaß, cf. APl 43. 3 (in the same metrical position); s0n kam1tzn, cf. AP I, 9. 1; kaò töde g2r t6menoß panao5dimon ™xet6lessaß, cf. AP I, 9. kaò töde s0n kam1tzn panao5dimon Çrgon ™t7cqh. Is this the work of a local poet? Perhaps, but given the superb literary quality of the verses it seems more likely that the palace official Leo the Protospatharios (the subject of the poem) commissioned a Constantinopolitan poet to compose this elegant verse inscription125. The second classicistic verse inscription is an early tenth-century epitaph found on a sarcophagus in the vicinity of Galakrenai, the monastery of the Patriarch Nicholas Mystikos126. The patriarch’s synkellos, Michael, is commemorated in the epitaph. The poem is remarkable for its use of Nonnian phrases, Homeric tags and explicit borrowings from the Greek Anthology. See, for instance, the following two macaronic verses (vv. 3–4): 4cqoß äporr5vaß (AP VII, 19. 4) bebarhöta (Homer and later epic writers) desmñn äl7xaß (Od. 8. 353) / possòn ™laórot1toisi (Nonnos, Dion. 28. 287, 32. 246, Par. Ev. Ioh. 19. 21) di6sticen (Nonnos, passim), ¼ci core7ei (Nonnos, Dion. 3. 110)127. Seeing that Alexander of Nicaea wrote two epitaphs on Nicholas Mystikos (APl 21–22), he would be a likely candidate if one desired to attribute this classicistic verse inscription to a known author; at any rate, the poem “emanated from (…) the same competent literary milieu of high prelates gravitating around the Great Church”128. However, it must be said that all this classicizing between c. 850 and 950 was very much a Constantinopolitan thing. The epigrams of the Anonymous Italian, for instance, are not at all classicistic. And even in Constantinople, the classicizing vogue was not wholeheartedly embraced by all intellectuals. The epigrams by Leo Choirosphaktes, for instance, are not particularly classicistic. Strangely enough, though, the same Leo Choirosphaktes was accused of “hellenism” by Arethas of Caesarea, an author whom we know to have written extremely classicizing epitaphs. The above is merely intended as a cautious reminder not to stick stylistic labels on periods. Diverging styles, preferences and mindsets coexist in Byzantium at any given moment, sometimes peace- fully, sometimes with a lot of sabre-rattling. No period is exclusively this or that. For instance, the art-historical concept of the “Macedonian Renaissance” may account for the classicistic style of the Paris Psalter, but ignores other,

125 See also the comments by OIKONOMIDES 1994: 489–492. 126 See ŠEVCENKO 1987: 461–468 and CAMERON 1993: 319. 127 See the excellent commentary by ŠEVCENKO 1987: 462 and 464. 128 ŠEVCENKO 1987: 462. Anthologies and Anthologists 121 non-classicizing styles such as oriental motifs on Byzantine silks129. Likewise, the hellenism of Leo the Philosopher and the christianized classicism of Photios and Cephalas should not be seen as the sole cultural forces in the ninth and early tenth centuries, but merely as determinant factors in an ongoing debate on Byzantium and the classical heritage. Debates are never won by any one party; at best the parties involved reach a meagre compromise, but if that is not possible, they keep on arguing for ever. Conflicts on the issue of hellenism kept flaring up in Byzantium from time to time, not because the Byzantines were constantly in some sort of identity crisis, but because they attempted time and again, with little success, to redefine the classical past in the light of their own experiences and needs130. Constantine the Rhodian annotated with obvious indignation at AP VII, 311: “on the wife of Lot, but the Hellenes say that it alludes to Niobe”. There can be little doubt that Cephalas is the target of criticism here, for the Planu- dean Anthology and the Sylloge Euphemiana, which both derive from the anthology of Cephalas, introduce the epigram as follows: “on Niobe”131. Constan- tine the Rhodian criticized Cephalas for failing to notice an obvious link with the biblical story of Lot’s wife turning into a pillar of salt. That the epigram obviously refers to the story of Niobe, was apparently of little concern to Constantine. In his view, it was a crying shame that Cephalas, who was to become prztopap@ß (would you believe it), did not draw the parallel with Lot’s wife where he easily could have done so. Constantine the Rhodian did not object to classical literature, of course, for otherwise he would not have invest- ed time and money in the compilation of what was to become the Palatine Anthology; but he certainly did not cherish an unreserved admiration for the classics. At AP VII, 26, a laudatory epigram on Anacreon, he wrote the following nauseated comment: “with filthy praises you crown a filthy man” – which clearly indicates that Constantine the Rhodian disapproved of Anacre- on’s poems on wine and women. But strangely enough, the same Constantine the Rhodian filled a whole quaternion of his own manuscript with various Anacreontea. What are we to make of this? It does seem quite schizophrenic to rebuke Anacreon first for his utter immorality and then publish the poems that go under his name. But if we could ask Constantine the Rhodian for his views

129 See H. BELTING, in: Byzanz und der Westen. Studien zur Kunst des europäischen Mittelalters. Vienna 1984, 65–83. 130 See H. HUNGER’s interesting comments on the Byzantine anthologists and the classical heritage, in: 17th International Congress of Byzantine Studies. Major Papers. Washing- ton 1986, 518–519. 131 The epigram is also mentioned in other sources. Eustathios at Il. 24. 614 and a scholion at Soph. El. 150 state that it refers to Niobe; Manuel Holobolos (ed. TREU 1893: 7) connects it with the story of the wife of Lot. 122 Part One: Texts and Contexts on the subject, I think that he would tell us that we really should learn to distinguish between form and content. Anacreon’s poems are distasteful, no doubt about that, but he writes excellent verses and we moderns can learn a great deal from him. His style is really superb. Don’t you recall that I, Con- stantine the Rhodian, used one of his impressive similes in my satire on wretched Theodore the Paphlagonian?132 Well, the same goes for all those ancient epigrams I copied myself or had copied by those scribes working for me. In many epigrams there is hardly anything I approve of, but let the truth be said: the ancients really knew how to write a poem. In AP XV, 17, probably the best epigram he ever wrote, Constantine the Rhodian states his views on art in unmistakable terms. It is an epigram on a picture of the Holy Virgin, in which Constantine considers the problem of representativeness. As he observes, the Holy Virgin rightfully deserves to be portrayed with stars and luminaries, but since that is clearly beyond our capacity, the artist has to content himself “with the material that nature and the laws of painting afford". Since literature and art are two forms of imagination that interact and respond to each other, especially in Byzantium where artists paint in words and write in paint, and since the epigram is as much an artefact as the picture it discusses, we may interpret Constantine’s words as his personal ars poetica. Poetry results from the lucky combination of sense and sensibility. That is to say, by observing the phenomena of nature and studying the rules of the art, a sensitive poet will learn how to write a good poem. But if he is intelligent enough, the accomplished poet will recognize the limitations of his art and will understand that there are things that cannot be fully expressed because they “do not yield to the voice of mortals”. He will know that subjects that transcend the human mind (such as the subject of the Holy Virgin) demand to be treated with the help of substitutes: symbols, circumlo- cutions and metaphors that indirectly reflect the reality of the supernatural, such as, for instance, the colourful expression uztñß p7lh used by Constantine to address the Holy Virgin. The book of nature provides the poet with all the images he needs and the books of the ancients instruct him how to use these images adroitly. However, if the poet were to use the symbols of imagination purely for art’s sake without referring to the divine secrets they reveal (as did Kometas in the eyes of Constantine), he would accomplish nothing. Reading the various poems of Constantine the Rhodian, there can be little doubt that he was well-read and knew both ancient and Byzantine poetry by heart. However, he never “classicizes”. He does not plagiarize ancient texts word for word, but merely selects expressions and images that fit into the context of the poem and are suited to convey the poetical message. Without Constantine’s

132 Ed. MATRANGA 1850: 628, v. 39. Cf. Anacreon 6, v. 7. Anthologies and Anthologists 123

Palatine Anthology we would know hardly anything about the hundred years of classicism, but Constantine the Rhodian himself had nothing to do with this cultural movement. In fact, he definitely was an exponent of Byzantine “modernism” – the feeling of being Byzantine and the tendency to articulate this feeling in ways that run counter to the stifling rules of classicism.

** *

The Anthologia Barberina

The history of the Greek Anthology from Leo the Philosopher to Constan- tine the Rhodian, as sketched in the above, would certainly present a distorted image of the cultural life in Constantinople in the years between 850 and 950, if people were to think that the key concept of classicism suffices to explain all the cultural phenomena of this period. For, as I stated previously, divergent styles and ideological preferences co-exist in Byzantium without any presump- tion to be mutually exclusive. In the following I shall discuss an early tenth- century anthology that is definitely not classicistic. Barb. gr. 310 is a small-size parchment manuscript of great beauty written in the second half of the tenth century133. The manuscript is extremely pre- cious, not only because of its elegant layout and handwriting, but also because of its contents. The manuscript used to contain a highly interesting collection of anacreontics and alphabets, which regrettably has not been preserved entirely because of the loss of some twenty-five quires. Fortunately, however, the index of the manuscript is still there to inform us what the manuscript contained before it was badly damaged. Some fifteen years ago the late Galla- votti produced an admirable edition of the index, together with a lucid and very learned commentary134. I follow his numbering and I use the name that he invented to christen the collection of anacreontics and alphabets: Anthologia Barberina (AB). The Anthologia Barberina is divided into two parts: nos. 1–80 and 81–160; the former contains anacreontics and the latter alphabets in accentual me- tres135. The layout of the two parts of the manuscript differs strongly. The alphabets are not written line by line, but continuously, without any regard for the metrical structure; the musical mode to which they are set is indicated in the manuscript and the names of the authors are written in the margin. The

133 See M.L. AGATI, Byz 54 (1984) 615–625 and 55 (1985) 584–588. 134 GALLAVOTTI 1987. 135 See GALLAVOTTI 1987: 60–70. See also CRIMI 2001: 28–51. 124 Part One: Texts and Contexts anacreontics, on the contrary, are written line by line, the musical mode is not recorded (with the exception of AB 24) and the names of the authors are mentioned in the titles attached to the poems. Whereas the index duly records the names of the authors of the anacreontics, it does not mention the authors of the alphabets by name, but rather niggardly introduces the second part as follows: älóabht1ria ×tera diaóörzn poiht0n, without telling us who these “various poets” are. That is a great pity, for almost all the poems of the second part have been lost, with the exception of AB 134–135 (by Christopher Prota- sekretis), AB 136–137 (by Photios) and AB 138 (anonymous)136. The first part of the Anthologia Barberina can be divided into five heteroge- neous sections: (a) the Palestinian school AB 1–22 Sophronios Patriarch of Je- rusalem AB 23 Sophronios Iatrosophistes AB 24–25 Elias Synkellos of Jerusalem AB 26 Michael Synkellos of Jerusa- lem (b) Constantinopolitan poets AB 27 Ignatios the Deacon AB 28–32 Arethas of Caesarea AB 33–38 Leo Choirosphaktes AB 39 Ps. Leo Choirosphaktes (c) sixth-century grammarians AB 40–46 John of Gaza AB 47–57 George the Grammarian (d) ninth-century grammarians AB 58–59 Leo the Philosopher AB 60 Sergios the Grammarian AB 61 Leontios the Grammarian AB 62–63 Constantine the Grammarian AB 64 Theophanes the Grammarian (e) Anacreon AB 65–80 In its present state the manuscript preserves only the following anacreontics: AB 1–13; the beginning of 14; the end of 16; 17–27; the end of 35; 36–45; the end of 49; 50–57. The following anacreontics can be found in other manuscripts: AB 14, 27, 52, 62–63 and 65–80. The second part of the Anthologia Barberina contains various hymns: pen- itential (nos. 93–123 and 146–154), on biblical and religious subjects (nos. 81– 87, 89–92 and 126–132), and ceremonial (nos. 88, 124–125, 133–145 and 155)137.

136 Christopher Protasekretis: ed. CICCOLELLA 2000b: 72–77; Photios and the anonymous poem AB 138: ed. CICCOLELLA 1998: 308–315. 137 AB 156–157 mention only the heirmos, not the subject. AB 158–160 are entitled nekrwsimon. Anthologies and Anthologists 125

The ceremonial hymns are poems that were performed at the imperial court in order to celebrate a certain historical event. The five ceremonial hymns that are still extant in the manuscript (nos. 134–138), celebrate emperor Basil I: poems AB 134–135 deal with Basil’s conversion of the Jews in c. 874, poems AB 136–137 refer to the council of 879–880 and Basil’s attempts to put an end to the discord between the Photians and the Ignatians, and poem AB 138 is an anthem performed at Basil’s coronation in 867. Some of the ceremonial hymns that are missing in the manuscript can be dated precisely: (139) a monody on the death of Basil’s son Constantine in 879, (140–141) monodies on the fall of Syracuse in 878, (142) a monody on the fall of Thessalonica in 904, (143–145) monodies on the death of Leo VI in 912, and (155) a poem on Andronikos Doukas’ revolt in 906–908. As we can see, all the datable poems in the second part of the Anthologia Barberina were composed in the short period between 867 and 912. The only exception to this rule is AB 88, “on Constantine the Emperor”. Likewise, none of the anacreontics found in the first part of the Anthologia Barberina were written after 912 (the death of Leo VI), again with one exception: AB 39. AB 39 is an epithalamium on the marriage of Constantine VII and Helen in 919. In the manuscript the poem is attributed to Leo Choirosphaktes, but it is beyond any doubt that the ascription is incorrect. The poet of AB 39 plagiariz- es Choirosphaktes’ epithalamium on the second marriage of Leo VI (AB 36) almost line by line; on the rare occasions that he attempts to produce a verse of his own, he commits prosodic blunders such as Choirosphaktes, a competent author, would never have allowed138. It is fairly easy to understand the error. As AB 39 follows immediately after other poems by Choirosphaktes (AB 33–38) and as it is just a cento of verses taken from an authentic epithalamium by Choirosphaktes, the scribe of Barb. gr. 310 quite understandably assumed that the poem should be attributed to the same Leo Choirosphaktes and therefore added the fateful words to¯ aJto¯. Since AB 39 is the latest datable poem of the collection of anacreontics and alphabets in Barb. gr. 310, it is reasonable to assume that the Anthologia Barberina was compiled in 919 or shortly afterwards. If the anthology had been compiled in the second half of the tenth century (the date of the manuscript), one would expect to find numerous anacreontics and alphabets written in honour of Constantine VII, Romanos II, and other members of the Macedoni- an dynasty, but this is not the case. As for the identity of the anthologist, I would suggest that he is the same person who wrote AB 39, which is the only anonymous poem in the first part of the anthology – anonymous precisely because the author and anthologist did not want to sign his own literary

138 See NISSEN 1940: 60–62, GIARDINA 1994: 9–22, and LAUXTERMANN 2003b. 126 Part One: Texts and Contexts composition out of pure modesty. The anthologist must have been a court dignitary of some importance, for he had access to the imperial archives, where the numerous hymns composed for performance at the imperial court were kept. For his anthology he selected only court poetry connected one way or another with the Macedonian dynasty. There are no ceremonial hymns in honour of Michael III or Theophilos, although they surely must have existed. The anacreontic part of his anthology is characterized by the same ideological bias. There is one poem on caesar Bardas (no. 58) and no less than seven poems on Basil I, Leo VI and Constantine VII (nos. 30–32 and 36–39). The pro– Macedonian orientation of the Anthologia Barberina139 strongly suggests that the anthologist wished to flatter the reigning emperor by including anacreontics and hymns celebrating his illustrious forebears. It is therefore very likely that the anthology was compiled in honour of, or perhaps even on behalf of, emperor Constantine VII. The anthologist may have presented the manuscript of the Anthologia Barberina, together with the poem he had written himself, to Constantine VII on the occasion of the emperor’s marriage to Helen Lekapene. The index of the Anthologia Barberina reads as a literary history in short. It rightly begins with Sophronios, the first practitioner of the Byzantine anacreontic. Then we have three Palestinian poets who followed in his footsteps: Sophronios Iatrosophistes140, Elias Synkellos141 and Michael Synkellos142. In the early ninth century the anacreontic left its native soil and was brought to Constantinople by Palestinian émigrés, such as Michael Synkellos. Ignatios the Deacon was the first Constantinopolitan to write anacreontics, just as he was the first poet to write classicizing elegiacs after c. 800. The compiler of the Anthologia Barberina then turns to the poets of his time: AB 28–39 are anacreontic compositions by Arethas, Leo Choirosphaktes and the anthologist himself. The next two sections in the Anthologia Barberina (40–57 and 58–64) are devoted to grammarians of the early sixth and the ninth centuries, respectively. It is worth noticing that all these poets are called grammatiköß, except for the arch-grammarian Leo the Philosopher. Thus the compiler of the Anthologia Barberina, whether correctly or not, connects these poets and their poems to the Byzantine school system. These products of the Byzantine classroom are followed by sixteen ancient Anacreontea, perhaps because they were read at school. These sixteen poems derive from a much larger collection of Anacre-

139 See CRIMI 2001: 46–53. 140 The author should not be confused with his more famous namesake: see TH. NISSEN, BZ 39 (1939) 349–350. Perhaps he is the Sophronios who used to teach in Edessa around the year 800: see A. MOFFATT, in: Iconoclasm, ed. A. BRYER and J. HERRIN. Birmingham 1977, 89, n. 32. 141 Elias Synkellos probably lived in the eighth century: see LAUXTERMANN 2003b. 142 On the life of Michael Synkellos (761–846), see CRIMI 1990: 5–11. Anthologies and Anthologists 127 ontea, probably dating from the sixth century, a copy of which is found in the Palatine manuscript143. Without the Anthologia Barberina we would know practically nothing about the history of the Byzantine anacreontic. Though he never inspected the manuscript, Nissen’s famous monograph on the Byzantine anacreontic is essentially a study of the Anthologia Barberina. It is an excellent account of the historical development of the anacreontic, but it could have been much better, had he studied the manuscript and its index instead of relying on unreliable editions (such as, notably, the Anecdota Graeca by Matranga)144. The Anthologia Barberina has little in common with the Greek Anthology. Whereas Cephalas collected ancient epigrams, AB is basically an anthology of Byzantine poems. Cephalas stops at c. 600 (with some exceptions); AB literally begins at c. 600 with the anacreontics of Sophronios. Cephalas includes the epigrams of Agathias and his circle because they clearly imitate Hellenistic models; but AB contains the poems of John of Gaza and George the Grammar- ian because they form the prelude to the Byzantine anacreontic. And while the Palatine manuscript contains the collection of Anacreontea in full, AB has only a mere selection. However, the most revolutionary aspect to the Anthologia Barberina is most certainly the inclusion of a large corpus of poems in accentual metres (the paired heptasyllable, the paired octosyllable, and probably also the political verse)145. These alphabets were added to the collection of anacreontics because both categories, alphabets and anacreontics, were intended for musical per- formance146. The Anthologia Barberina is in fact a collection of lyrics. It is a songbook without musical notation. The only parallel to this songbook in tenth-century Byzantium is the famous Book of Ceremonies, where we also find numerous librettos with hardly any indication of how these acclamations may have sounded147. However, whereas the Book of Ceremonies contains texts for

143 See M.L. WEST, Carmina Anacreontea. Leipzig 1984, X–XI. 144 All the poems in Barb. gr. 310 have now been edited properly: GIGANTE 1957, CRIMI 1990, and CICCOLELLA 1998, 2000a and 2000b. But we still need a comprehensive edition of the Anthologia Barberina, including the index, all the poems still extant in the manuscript as well as the poems that are no longer there, but which can be found in other manuscripts. 145 See LAUXTERMANN 1999c: 48–51. 146 For the musical performance of the alphabets, see the lemmata attached to AB 134–138. Zonaras, Life of Sophronios (see NISSEN 1940: 5, n. 2), informs us that Sophronios’ anacreontics were meant to be sung; cf. the title of Elias’ anacreontic, AB 24; see P. SPECK, Das geteilte Dossier. Bonn 1988, 364–365. 147 For a metrical analysis of some of the acclamations, see LAUXTERMANN 1999c: 61–65. For the musical performance of the acclamations, see J. HANDSCHIN, Das Zeremonienwerk Kaiser Konstantins und die sangbare Dichtung. Basel 1942 and E. WELLESZ, A History of Byzantine Music and Hymnography. Oxford 21961, 98–122. 128 Part One: Texts and Contexts recurrent festive occasions, most poems in the Anthologia Barberina were composed for a one-off event. Is the purpose of AB “antiquarian”? In various scholarly publications Constantine VII is praised for, or accused of, his alleged “antiquarianism” – which is rather an unlucky catch phrase to denote the various cultural phenomena of his long reign. The Anthologia Barberina is perhaps “antiquarian” inasmuch as it contains many poems that were composed for a specific moment in the past. But it is equally “modern”, as it provides models to be imitated for future occasions, such as the epithalamium on Leo VI (AB 36), which was re-used and adapted some twenty years later for the wedding of Constantine VII and Helen Lekapene (AB 39). More important- ly, however, an anthology containing a large amount of poems in accentual metres is really without precedent in the ninth and early tenth centuries. It is precisely for this reason that the Anthologia Barberina should be viewed as a novelty rather than as a supposedly “antiquarian” enterprise. Seen from the viewpoint of tenth-century Byzantium, the Anthologia Barberina opens up new perspectives on the recent, but somehow ever distant past.